Southern Blue Ridge: An Analysis of Matrix Forests - Conservation ...

Southern Blue Ridge:An Analysis of Matrix ForestsPhoto by Witt LangstaffProduced by The Nature ConservancyMark Anderson, David Ray, John Prince, Megan Sutton, and Angela WatlandDate of Analysis: 2010-2011Date of Report: April 2012

Suggested Citation: Anderson, M., Prince, J., Ray, D., Sutton, M., and Watland, A. 2013. Southern BlueRidge: an Analysis of Matrix Forests. The Nature Conservancy. pp. 51.http://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspxAcknowledgementsWe would like to thank the Glass Foundation for directly funding the expert meetings that were held inall five states of the Southern Blue Ridge and for funding the Forest Block video that was completed(http://www.youtube.com/watch?v=qMvwbaU5RWs).We would like to thank the numerous TNC staff that participated in this analysis including Kristen Austin,Catherine Burns, Deron Davis, Colette DeGarady, Sarah Fraser, Sara Gottlieb, Wade Harrison, MalcolmHodges, Eric Krueger, Merrill Lynch, Katherine Medlock, Sally Palmer, Marek Smith, Rick Studenmund,Randy Tate, Maria Whitehead, and Alex Wyss.We would also like to thank our partners that participated in this process including Jon Ambrose, JoanneBaggs, Rob Baldwin, Allen Boynton, Kevin Caldwell, Geoff Call, Ed Clebsch, Jamey Donaldson, MarshallEllis, Tom Govus, Mark Hall, Steve Hall, Dennis Herman, Stan Hutto, Hugh Irwin, Gary Kauffman,Christine Kelly, Josh Kelly, Nathan Klaus, Tim Lee, Jay Leutze, Jeff Magniez, Joe McGuiness, PatrickMcMillan, Joe Miller, James Padgett, Tom Patrick, Mark Pistrang, Ed Pivorun, Joe Pollard, Mike Schafale,Ed Schwartzman, Vic Shelburn, Curtis Smalling, Tom Waldrop, Jesse Webster, Kendrick Weeks, CarolynWells, Jim Wentworth, Jason Wiesnewski, Lori Williams, Claiborne Woodall, and Pete Wyatt.i

TABLE OF CONTENTSEXECUTIVE SUMMARY .............................................................................................................................1AN APPROACH TO LASTING FOREST CONSERVATION IN THE SOUTHERN APPALACHIANS ..........................................2OVERVIEW: THE SBR MATRIX FOREST BLOCK ANALYSIS PROCESS ......................................................................5DELINEATING MATRIX FOREST BLOCKS ........................................................................................................6SCREENING MATRIX FOREST BLOCKS FOR SIZE AND CONDITION .........................................................................8LAND COVER CRITERIA ..........................................................................................................................................................8SIZE CRITERIA ......................................................................................................................................................................8ELEVATION .......................................................................................................................................................................13GEOLOGY .........................................................................................................................................................................14ECOLOGICAL LAND UNITS ....................................................................................................................................................15EVALUATING AND PRIORITIZING A NETWORK OF MATRIX FOREST BLOCKS ......................................................... 22RESULTS OF ANALYSIS ........................................................................................................................... 23SIZE OF MATRIX FOREST BLOCKS .............................................................................................................. 23CONSERVATION STATUS AND CONDITION OF MATRIX FOREST BLOCKS .............................................................. 24LITERATURE CITED ................................................................................................................................ 29ii

Table 1. Data used to evaluate potential fragmenting features in matrix forest block delineation. ........... 7Table 2. Average Territory Size of Characteristic SBR Forest Birds ............................................................. 11Table 3. Ecological Land Units Coding Scheme. .......................................................................................... 16Table 4. Tier 1 Matrix Forest Block Size Cohorts and ELU Groups Represented. ....................................... 23Table 5. Tier 2 Matrix Forest Block Size Cohorts and ELU Groups Represented. ....................................... 24Figure 1. Southern Blue Ridge Ecoregional Boundaries. ............................................................................... 2Figure 2. Regional Forest Conservation Model.. ........................................................................................... 5Figure 3. Minimum Dynamic Area for SBR Matrix Forest Blocks. ............................................................... 12Figure 4. An example of landforms ............................................................................................................. 15Figure 5. Southern Blue Ridge Potential Matrix Forest Blocks and Ecological Land Units. ........................ 19Figure 6. Southern Blue Ridge Early Matrix Forest Blocks by similar Ecological Land Unit Groupings. ..... 20Figure 7. Prioritized Southern Blue Ridge Matrix Forest Blocks. ................................................................ 21Figure 8. SBR Matrix Forest Blocks Overlapping GAP Status. ..................................................................... 26Appendix A: State Expert Review Teams and TNC Analysis Team Members ............................................. 33Appendix B: Fragmenting Features: Detailed GIS info ............................................................................... 35Appendix C: Southern Blue Ridge Geology Classes..................................................................................... 36Appendix D: Matrix Forest Block Information ........................................................................................... 38Appendix E: Final Southern Blue Ridge Matrix Forest Block Statistics ....................................................... 41Appendix F: Key to Columns in Appendix E ............................................................................................... 48iii

EXECUTIVE SUMMARYThe Southern Blue Ridge (SBR) Ecoregion’s forested landscape (portions of Georgia, North Carolina,South Carolina, Tennessee, and Virginia) is comprised of intact temperate forest over a diversity oflandforms, elevation zones, and bedrock geologies, making it one of the most biologically-diverse areasin North America. This region contains several of the few remaining mega-blocks of relativelyunfragmented forest in the eastern United States, supporting the highest diversity of salamanders in theworld, a tremendous diversity of tree and herbaceous species, and very high densities of forest breedingbirds. These large contiguous forests provide fundamental ecosystem services that sustain underlyingnatural processes, ensuring the continued persistence of plant and animal populations as well as theprovisioning, regulating, and cultural ecosystem services on which humans depend (e.g., quality drinkingwater, flood control). From a global perspective, the Southern Blue Ridge forested landscape is a hugeand irreplaceable ecosystem recovering from regional-scale deforestation. These reestablished forestsare facing compounding and interacting threats due to increased human population, forestfragmentation, pests and pathogens, soil acidification and global climate change.Previous efforts by The Nature Conservancy (TNC) and its partners identified priority SBR conservationlocations, focusing largely on occurrences of rare species and communities at the scale of individualpatches; however, the scope and magnitude of today’s conservation challenges mean that we mustexpand our focus to include landscapes and strategies beyond a protected network of preserves. TheConservancy’s approach to the long-term conservation of the Southern Blue Ridge critical forestresource envisions the conservation of a connected, representative, well managed, matrix forests,embedded with large areas of core forest(s). The matrix forest blocks sustain natural cover throughmultiple-use working forests, while the core forests are protected and managed for natural ecologicalfunction that promotes biodiversity and natural disturbance regimes (i.e., a dynamic mosaic of stands indifferent age and seral classes).This report summarizes the process and results of the Southern Blue Ridge Matrix Forest Analysis,completed in 2011, which identified a representative network of matrix forest blocks, large andcontiguous enough to maintain key ecosystem processes and services, resilience, and movement oforganisms, and to provide for accommodation of catastrophic disturbances and the breeding needs offorest interior species. In 2011, TNC staff from Eastern North America Division Science and five stateoperating units (with significant contributions from several state and regional partners) completed afour-step analysis process to identify priority SBR matrix forests, involving (1) delineating matrix forestblocks (discrete blocks of contiguous forest, using roads and other fragmenting features in GIS), (2)screening each matrix forest block for size and condition using land cover and size criteria, related todisturbance and species’ needs, (3) classifying the matrix forest blocks into representative forestlandscape types, using elevation, geology and landforms (Ecological Land Units), and (4) evaluating andprioritizing a network of functional matrix forest blocks representative of the diversity of ecoregionalforest landscape types, using additional data and expert review.The results give us a much clearer understanding of the status, distribution, and spatial context of large,contiguous matrix forests in the Southern Blue Ridge, and will provide a clearer direction for near-termconservation priorities and foster a more focused set of conservation actions around which TNC andpartners can organize and cooperate. The Nature Conservancy’s vision for the conservation of theseidentified priority SBR matrix forest blocks is to work with partners to (1) ensure adequate long-term1

protection and ecologically-compatible management practices, (2) retain connectedness of forest coveramong and between them, and 3) work to abate region-wide forest threats.AN APPROACH TO LASTING FOREST CONSERVATION IN THE SOUTHERN APPALACHIANSIf one imagines oneself looking out a plane window on a flight from Atlanta, GA toward Pittsburgh, PA, alarge-scale perspective on the Southern Blue Ridge (SBR) Ecoregion can be visualized (Figure 1). At thiselevation, some of the few remaining mega-blocks of relatively unfragmented forest in the easternUnited States are readily apparent toeven the untrained eye. For instance,10 of the matrix forest blocksidentified in this study are between100,000 and 300,000 acres in size.These forests are important for thepeople inhabiting the region, whodepend on them not only forprovisioning and regulating ecosystemservices (e.g., quality drinking water,natural resource extraction and use,erosion and flood control) but also forcultural services (i.e., the non-materialbenefits and renewal that comes frombeautiful scenery and recreation). Ofcourse, these large contiguous forestsalso provide fundamental ecosystemservices that sustain underlyingnatural processes, ensuring theFigure 1. Southern Blue Ridge Ecoregional Boundaries.continued persistence of plant and animal populations as well as the provisioning, regulating, andcultural ecosystem services on which humans depend. The driving force behind these benefits to peopleand the reason the Southern Blue Ridge elicits both scientific and spiritual awe is this ecoregion’s wellrenownedglobally-significant biodiversity (e.g., unique natural communities like globally-rare grassybalds, Carolina hemlock bluffs, and southern Appalachian bogs).The Southern Blue Ridge Ecoregion means many things to many different people. Its 9.4 million acresare home to approximately two million people (US Census 2010), composed of some families that havebeen rooted in the land for generations, and others, lured largely by quality-of-life factors, that haverecently migrated there. Countless other part-time residents and visitors appreciate the beauty andrecreation of the Southern Blue Ridge mountains, valleys, and coldwater streams, including enjoyingscenic vistas along the Blue Ridge Parkway and within the Great Smoky Mountain National Park, as wellas fishing, hunting, hiking, and biking on 3.2 million acres of National Forest. Natural beauty and bountymeet cultural richness in small-to-mid-sized cities like Asheville, NC, Johnson City, TN, and Roanoke, VA,as well as in rural pockets of artistic expression like Yancey County, NC, and Blue Ridge, GA. Theheadwaters of the Southern Blue Ridge also influence several large cities, as the sources of drinking andrecreational waters for Atlanta, GA, Charlotte, NC, Knoxville, TN, and Greenville, SC. Research showsthat one-third of the world’s 105 largest cities’ drinking water sources arise from protected areas,demonstrating the benefit of well-managed natural forests to urban populations (Dudley and Stolton2003).2

The Southern Blue Ridge is a forested landscape of steeps slopes, high mountains, deep ravines, andwide valleys. The combination of intact temperate forest over a diversity of landforms, elevation zones,and bedrock geologies, makes it one of the most biologically diverse areas in North America. “The regionsupports the highest diversity of salamanders in the world, extremely rich forests with a tremendousdiversity of tree and herbaceous species, and very high densities of breeding birds” (Hunter et al. 1999).Five broad forest types, characteristic of the region, form the dominant matrix (Appalachian oakhardwoods, dry oak pine, cove forests, northern hardwoods, andspruce-fir), and their distribution tracks change with elevation. Otherforest types (such as riparian forests) occur at a smaller scale, usually inconjunction with a landform or specific setting. At the scale of largeintact forest areas (5,000 – 50,000 acres), the individual forest typesblend and intermix to form a larger functioning unit that shares manyprocesses and exhibits structural and compositional heterogeneity. Eachforest type will display a range of successional classes, given the abilityof various processes and disturbances to play out across the landscape.From a global perspective, the temperate deciduous and mixed forest ofthe Southern Blue Ridge is a huge and irreplaceable ecosystemrecovering from regional-scale deforestation. The great majority of thecurrent forest is mid-to-late successional; however much of its speciesand structural composition has been altered by past land uses andpractices such as agriculture, pasturing, fire suppression, and logging.Photo by Hugh MortonConcurrent with forest re-establishment, the human population hasincreased exponentially, leading to increased densities of roads and other urban environments whichhave led to greater forest fragmentation. Other stresses have expanded to include and compoundhabitat fragmentation, and threats such as increased pests and pathogens, soil acidification and globalclimate change continue to increase.In 2000, The Nature Conservancy (TNC) and its partners (Southern Appalachian Forest Coalition andAssociation for Biodiversity Information) conducted an Ecoregional Assessment for the Southern BlueRidge, identifying priority conservation locations (portfolio sites) known to harbor conservation targets(i.e., globally-rare plants, animals, and natural communities) (TNC & SAFC 2000). The EcoregionalAssessment, however, did not include identification and evaluation of large contiguous forested habitatsthemselves for conservation priorities, which form the very matrix supporting embedded conservationtargets. The scope and magnitude of today’s conservation challenges mean that we can no longerafford to limit our strategies to protecting a network of preserves, but must consider strategies andlandscapes large enough to maintain key ecosystem processes and services, resilience, and movementof organisms. TNC has recognized the need for this new “whole system” approach that involves workingat multiple scales, with an increasing emphasis on managing for connectivity and a permeable matrix oflands and waters that vary in quality, surrounding portfolio sites of high ecological integrity. In theSouthern Blue Ridge, this means looking across the landscape at large blocks of native habitat that cansupport the species, communities, and ecosystem processes and functions that will protect biodiversityand support people’s well-being now and into the future.The first iteration of the Southern Blue Ridge Ecoregional Assessment in 2000 (TNC & SAFC 2000)focused largely on rare species and communities or elements of biodiversity that occurred at the scale3

of individual patches. This follow up study was designed to specifically spatially-define the contiguous,large matrix-forming forest types (herein, “matrix forest”) across the landscape (Anderson 2008).Developed by TNC operating units in the northern Appalachians, this new assessment for the southernsections of the Appalachian Mountains has resulted in a much clearer understanding of the status anddistribution of matrix forest conservation targets in this ecoregion. This report presents the process,methods, and results of the analysis, completed in 2011. It closes a significant gap in our understandingof the spatial context and characteristics of large contiguous forests in the Southern Blue Ridge, and willprovide a clearer direction for near-term conservation priorities, and foster a more focused set ofconservation actions around which TNC and partners can organize and cooperate.Scientists have wrestled with how to conserve such a critical but compromised forest resource. TheNature Conservancy’s vision for the conservation of this forest includes employing strategies to (1)ensure adequate long-term protection and management of a network of representative matrix forestblocks, embedded core forests surrounded with working forest buffer (Noss and Cooperrider 1994,Figure 2), (2) retain connectedness of forest cover among and between those matrix forests and acrossthe landscape, and 3) abate region-wide forest threats. The identification of a representative networkof matrix forest blocks large enough to sustain forest ecosystem processes is the subject of this planningeffort (Anderson 2008).The Conservancy’s approach to the long-term conservation of the Southern Blue Ridge critical forestresource envisions the conservation of a connected, representative, well managed, matrix of forest,embedded with large core forest areas. The matrix sustains natural cover through multiple-use workingforests, while the core forests are protected and managed for natural ecological function that promotesbiodiversity and natural disturbance regimes. The desired condition of embedded core forest is not asolid stand of old-growth trees, but rather, a mosaic of stands in different age classes, from early to latesuccessional, that retain biological legacies (such as coarse woody debris and standing snags), andreflect the natural disturbance regimes of the region. This assessment describes the delineation ofmatrix forest areas large enough to provide for accommodation of catastrophic disturbances and thebreeding needs of forest interior species. Our methods anticipate forest matrix blocks that include arange of habitats and a shifting mosaic of seral stages (see section on Size in Relation to Disturbances).4

Figure 2. Regional Forest Conservation Model (adapted from Noss and Cooperridge 1994). Thisdisplays the concept of a model regional reserve network, consisting of core forests, connectingcorridors or linkages, and multiple-use buffer zones. “Matrix forests” refer to the overall contiguousforested landscape, while “core forests” refer to areas managed for natural ecological integrity.Inner buffer zones would be strictly protected, while outer buffer zones would allow for a widerrange of compatible human uses. The inter-regional corridor connects this system to a similarnetwork nearby.OVERVIEW: THE SBR MATRIX FOREST BLOCK ANALYSIS PROCESSThe Southern Blue Ridge Matrix Forest Block Analysis was a joint effort among TNC’s Eastern NorthAmerica Division Science staff and staff from the five TNC state operating units partially within theecoregion (Georgia, North Carolina, South Carolina, Tennessee, and Virginia). Significant contributionsand feedback were also provided by several state and regionalpartners (Appendix A). This process was similar to theConservancy’s matrix forest analysis in the CentralAppalachian and Northern Appalachian Ecoregions tofacilitate regional consistency in the identification andevaluation of Appalachian spatial conservation targets andecological goals. The major steps and details of the analysisprocess are summarized in this report, and supportingdocuments and additional details can be found at:http://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspx.5

We developed a four-step analysis process to assess, prioritize, and direct conservation strategies toprotect the Southern Blue Ridge matrix forest system:• Delineate matrix forest blocks to identify and subdivide the forested landscape into discrete blocksof contiguous forest, using roads and other fragmenting features in GIS.• Screen each matrix forest block for size and condition using land cover and size criteria, related todisturbance and species’ needs.• Classify the candidate matrix forest blocks into multiple groups that reflect similar combinations ofelevation, geology, and landforms (Ecological Land Units).• Evaluate and prioritize a network of functional matrix forest blocks representative of the diversityof ecoregional forest landscape types, using additional data and expert review.DELINEATING MATRIX FOREST BLOCKSUsing GIS data, we subdivided the entire forested landscape of the Southern Blue Ridge into discreteunits (“matrix forest blocks”), bounded on all sides by major roads (road classes 1-4; ESRI 2009), lakes,or large rivers (>3,681m 2 drainage area) (USEPA and USGS 2008) (Table 1 and Appendix B). Roads werean appropriate choice for this task, as they disrupt the movement of some organisms and the flow ofsome ecological processes, and they increase the level of access into interior forest areas. Additionally,we analyzed and sometimes adjusted blocks according to the location of roads, railroads, powerlines,logging trails, housing developments, agricultural lands, and mining operations from obtained digitaldata or aerial imagery (Table 1) that can be highly correlated with human natural resource extractive orother ecologically-incompatible activities that can impair connectivity and forest processes. DelineatedSBR matrix forest blocks were bounded on all sides by connected fragmenting roads (or other features),forming polygons that completely encircled large contiguous patches of forest. However, delineatedblocks are not necessarily roadless, as blocks can contain intersecting roads (roads that extend intoblock boundaries, but do not bisect the block).6

Table 1. Data used to evaluate potential fragmenting features in matrix forest block delineation.PotentialFragmentingFeatureRoadsData Analyzed(Year)Streetscarto layerfrom ESRI GIS data(ESRI 2009)Description of AnalysisDetailed road data was evaluatedagainst aerial imagery (accessedusing Google Earth and ArcGISOnline 2010-2011) to determineaccuracy and completeness of roadcoverage.Final Decision: FragmentingFeature?Road classes 1 through 4, which roughlyequates to Interstates (1), U.S. Routes andsome State Routes (2), State Routes and oldhighways (3), and exit and on ramps, serviceroads, and rest area roads (4), weredetermined to be fragmenters.PowerlinesRailroadsVentyx powerlinetransmission data(Ventyx 2010)Ventyx Railroaddata (Ventyx 2010)Powerline Transmission data wasevaluated against aerial imagery(accessed using Google Earth andArcGIS Online 2010-2011) to assessvegetation cuts (width andmanagement intensity).Railroad corridors were evaluatedagainst aerial imagery (accessedusing Google Earth and ArcGISOnline 2010-2011) to assessvegetation gaps (width,management, tunnel openings, andadjacency to roads/ rivers).Due to high variability of the width of thepowerlines and the degree to which theywere maintained as cut areas across theregion, powerlines were not uniformlyviewed as fragmenting features but wereconsidered on a case-by-case basis; wherepowerlines were large in width and requiredvegetation management continuouslythroughout individual blocks, these wereviewed as fragmenters.Due to sporadic traffic and often narrowopenings made for trains, railroads were notviewed as uniform fragmenting features;railroad attributes (width, management,tunnel openings, and adjacency toroads/rivers) were considered as potentialfragmenters within individual blocks.LakesNationalHydrographyDatabase (USEPAand USGS 2008)Waterbodies were evaluated todetermine if they created significantgaps in forest canopy.Determined to be fragmenting features.RiversNationalHydrographyDatabase (USEPAand USGS 2008)Waterbodies were evaluated todetermine if they created significantgaps in forest canopy.Large rivers or stream segments (draining>3,861 miles 2 ) were determined to befragmenting features.Developed/DisturbedAreasSE GAP NLCD 2006land cover (Fry etal. 2011)Land cover data was utilized tovisually assess fragmentation offorest cover.Not used as fragmenting features duringinitial block identification. Somemodifications to blocks were manually madebased on this later in the process.7

SCREENING MATRIX FOREST BLOCKS FOR SIZE AND CONDITIONOur work to identify a network of representative forest reserves posed several important questionsrelated to the size and condition of matrix forest blocks: How large must a forest reserve be to remainresilient in the face of a changing climate? How large must it be to contain all of its expectedbiodiversity? How can we tell if a forest occurrence has integrity with regard to its internal processes?In this section, we describe the criteria developed to identify qualifying matrix forest blocks in thisanalysis.Our goal was to identity matrix forest blocks large enough to function as viable, resilient forestecosystems. At a minimum, we assumed that a viable matrix forest patch includes both the biota andthe functions that arise from their interactions (e.g. pollination, decomposition), and the intactness ofthe physical setting and external processes that sustain the ecosystem (Forman 1995, Franklin 1993).Further, our working definition of resilience was the ecosystem’s “capacity to renew itself or adaptwithin a dynamic environment” (Gunderson 2000). We assumed that viable, resilient systems are morelikely to persist, not in a static manner but in a dynamic state that fluctuates within some bounds ofvariation or that adjusts gradually to new situations if the underlying environmental conditions change.Under a changing climate, the individual species that compose the system may change in abundance,and new species may be introduced to the system, but overall the forest continues to support a diversityof species and maintain its essential processes.LAND COVER CRITERIATo ensure that this analysis focused on the identification of contiguous forest, we used the SE GAP landcover GIS database (2008) to tabulate the amount of forest cover within each potential matrix forestblock. All blocks containing at least 80% forest cover qualified for further review.SIZE CRITERIAThe ability of a forest patch to provide adequate breeding territories for multiple pairs of forest interiorspecies, and to remain resilient under a suite of expected large disturbances, is correlated with its size(Poiani et al 2000, Anderson 2008). To understand the matrix forest block size necessary for effectiveforest conservation in this region, we identified two independent sets of size criteria, using (a) the scaleand frequency of natural disturbances to estimate the minimum dynamic area, and (b) the breedingarea requirements of interior-forest specialist bird species, to ensure that conserved matrix forests areof adequate size to function as a “coarse-filter” habitat for associated species, accommodating a broadrange of seral stages, at different elevations (Hunter 1996).Size in Relation to DisturbanceTo persist over time, a forest block must be large enough to absorb large, infrequent, disturbances.Natural disturbances, although catastrophic relative to the mortality of living trees, ultimatelyrejuvenate ecosystems by releasing and redistributing resources, and creating successional habitat thatfavors a different set of species. Spatial variation in disturbance severity and frequency breakshomogeneous areas into a mosaic of overlapping heterogeneous patches, and keeps the system in adynamic state of flux. The area necessary to maintain processes and ensure persistence has been calledthe system’s minimum dynamic area (Pickett and Thompson 1978).Primary natural disturbances in the Southern Blue Ridge are fire, drought, floods, tornadoes, hurricanes,ice storms, landslides, pathogens, and insects. Some sites, because of their position along environmental8

gradients, are more prone to certain disturbances. For example, ice damage occurs more frequently atcooler and higher elevations, wind-throw is more likely on exposed slopes, drought mortality is morefrequent on ridge sites, and fires are more likely on dry, warm, low-elevation southern exposures.To estimate the minimum dynamic area needed to absorb the largest likely disturbance over the courseof several centuries, we collected and reviewed published information and expert knowledge on thelargest known disturbance patch sizes created by wind and fire in the SBR. We multiplied thesedisturbance patch sizes by four, theorizing that resilience would be enhanced if less than one-quarter ofthe forest block was in the immediate stages of recovery at any one time (Anderson 2008). Weemphasize that this is a ballpark estimate, as there is much more detailed research underway to identifyoccurrence probabilities for forest patch sizes of varying disturbances. Ideally, one would model variousdisturbances in specific areas to extract the maxima and minima and other distributional information (D.Loftis pers. com.). By scaling to the four-times the largest recorded damage patch, we attempt toidentify a robust size criterion to allow for a variety of scenarios to play out.In the Southern Blue Ridge, wind disturbance (tornadoes, hurricanes, and downbursts) primarily createssmall gaps on the scale of 0.1 acre (5-6 m 2 ) and ranging up to 2.7 acres (0.2-1.1 ha). New gaps form at anaverage rate of 1% of total land area per year, and in sample areas covered 9.5% of the land area(Runkle 1981, Runkle 1982, Greenberg and McNab 1998, Ventyx 2010). A recent tornado touched downin the western tip of the Great Smoky National Park in April of 2011, creating a disturbance patch 11.5miles long and one-quarter mile wide, equivalent to 2,080 acres in size (D. Ray, pers. comm.). Using thisinformation, the estimated minimum dynamic area (four-times largest known disturbance size) wouldbe 0.4 acres for small wind disturbance gaps, 10.8 acres for hurricane downbursts, and 8,320 acres fortornado damage patches in the Southern Blue Ridge.The role of fire in the Southern Blue Ridge is not well understood, but it is likely that the fire regimes ofthe mountains are distinct from the well-studied Piedmont and Coastal Plain. There is evidence that fireshave regularly burned across some mountainous areas over the past 4,000 years, although theseasonality and severity of these fires remains unknown (Fesenmyer and Christensen 2010). Recentstudies in the nearby Central Appalachians found the average disturbance size of 344 natural burns was3 acres (1.2 ha) and the largest known natural fire was 2,938 acres (1,189 ha, Lafon et al 2005). In astudy on Tennessee and North Carolina forest fires, the average disturbance size of 126 lighting-causedfires was 2 acres (8 ha), with the largest lightning fire on record burning 81 acres (33ha) (Barden andWoods 2008). Evidence suggests that fires throughout the Southern Blue Ridge similarly occur, mainly aslow intensity burns. Scaling up to four-times the largest known natural fire damage area, the estimatedminimum dynamic area would be 11,752 acres for fire disturbances.Size in Relation to SpeciesTo understand the area needs of species associated with Southern Blue Ridge forests and forest interiorhabitat, we identified a set of species typical of, or restricted to, this ecoregion and used availableinformation on home range, resource, or breeding habitat needs to determine the minimum forest sizearea requirements. We restricted the analysis to vertebrates, as they are usually the most spacedemandingand wide-ranging, and therefore assumed that the space requirements of smaller specieswould be met at sizes sufficient for representative vertebrates. We focused on bird species requiring orassociated with forest interior habitat, as these species have the most restrictive requirements withregard to forest size (Martin and Finch 1995). The effects of forest patch size and forest fragmentationon particular species (specifically on migratory songbirds) have received extensive attention in the9

literature, and it is well documented that the density of many neotropical forest dwelling bird species isgreater in large forested tracts than small forested tracts (Robbins et al. 1989, McIntyre 1995). Thedifferences may be due to: increased pairing success, more suitable foraging or nesting sites, decreasedcompetition for resources, decreased nest predation and/or decreased nest parasitism (Paton 1994,Hartley and Hunter 1997, Fahrig 2003).The Southern Blue Ridge ecoregion supports an estimated 155 species of breeding birds (Hunter et al.1999). Nearly 62% of all these species are associated with forest habitats, ranging in size from smallwoodlots and groves at lower elevations, to large, extensively forested tracts at mid to high elevations.Low elevation forests support six species of woodpeckers, along with many songbird species. Forestpatches at higher elevations contain species that use mid-to-late- successional spruce-fir forests asbreeding or foraging habitat (Hamel 1992). Hunter et al. (1999) estimated that 49 species of neotropicalmigrant birds are forest-dependent, and 14 species are associated with wetland or riparian habitats.Population declines have been evident since 1996 in nearly 70% of nearctic-neotropical migrant birdspecies that breed in mid-to-late successional forests, while 26% have declined significantly. In addition,31% of wetland and riparian neotropical migrants have also declined (Hunter et al. 1999), andpopulations of most disturbance-dependent bird species (e.g., golden-winged warbler) in eastern NorthAmerica have declined steeply (Hunter et al. 2001).Breeding territory sizes, home ranges, and resource needs for characteristic forest-interior specialist birdspecies were compiled from Birds of North America (Poole and Gill 1992-ongoing). To determine the sizeneeded for matrix forest block occurrences to contain multiple populations and potentially function as asource area for breeding species, we multiplied the average female territory size by 25, using theguideline that at least 50 genetically-effective individuals are necessary to conserve genetic diversitywithin a metapopulation over several generations (Lande 1988). In using this guideline, we wereapproximating the area needed to accommodate 25 breeding females within a single patch of forest,not the needed population size to sustain the species, which would be much larger. Results show thatthe territory sizes of SBR forest interior bird specialists ranged from 2 acres to over 1000 acresdepending on the species (Table 2).10

Table 2. Average Territory Size of Characteristic SBR Forest Birds(compiled by M. Lynch and supplemented by Poole et al. 2012).AverageSouthern Blue Ridge Birds 1 Territory(Acres)Acres X 25Early SuccessionalGolden-winged Warbler 2 50Vesper Sparrow 10.4 260Cove HardwoodsAcadian Flycatcher 2.7 68Black-throated Blue Warbler 2 50Cerulean Warbler 2.6 65Louisiana Waterthrush 1.3 33Swainson's Warbler 2 50Hemlock-White Pine-Mixed HardwoodsBlackburnian Warbler 3 75Black-throated Green Warbler 1.5 38High Elevation HardwoodsRuffed Grouse 5.4 135Yellow-bellied Sapsucker 14 350Black-billed Cuckoo 15 375Blue-headed Vireo 0.7 18Rose-breasted Grosbeak 2.5 63Veery 0.5 13Spruce-FirBlack-capped Chickadee 10 250Brown Creeper 11 275Northern Saw-whet Owl 247 6175Red-breasted Nuthatch 12.6 315Golden-crowned Kinglet 2 50Winter Wren 7.2 180Large TerritoryBarred Owl 600 15,000Common Raven 1,200 30,000Cooper's Hawk 500 12,500Broad-winged Hawk 569 14,224To determine the matrix forest block size necessary for effective forest conservation in this region, weplotted these SBR forest-interior species’ area requirements on the same scale as the collecteddisturbance patch minimum dynamic area estimates (Figure 3). For instance, a 1,000 acre patch of forestprovides adequate space for 25 breeding pairs of yellow-bellied sapsuckers and red breasted warblers,1 Additional requirements beyond size exist for breeding success for many birds, especially early successionalspecies.11

ut is below the size needed for northern saw-whet owls, and is less than half the size of the largestsevere fire recorded in the area.In reviewing these results, it was evident that a 15,000 acre forest would be adequate for all the speciesand disturbances examined. We therefore decided to use 15,000 acres of contiguous forest as our basesize criteria, but to still evaluate blocks in the smaller size range of 10,000-15,000 acres, because weexpected that large blocks may not be available in certain environments (rich soils for example) and the10,000 acre block size would still be adequate for most disturbances and species (in the end all the Tier1 blocks were greater than 15,000 acres in size but some of the Tier 2 blocks and connectors weresmaller). Thus our minimum size criterion for viable SBR matrix forest blocks was 15,000 acres. This sizewill provide adequate breeding territories for multiple pairs of forest interior species, and can remainresilient under a suite of expected severe wind and fire disturbances of all types.Figure 3. Minimum Dynamic Area for SBR Matrix Forest Blocks, known disturbance patch sizesand select interior-forest breeding specialists’ territory sizes. The blue circle indicatesapproximately where the 15,000 acre criteria stand in relation to the other features.12

CLASSIFYING THE MATRIX FOREST BLOCKSAn understanding of patterns of environmental variation and biological diversity is fundamental toconservation planning at any scale—regional, landscape level, or local. Moreover, maintaining forestbiodiversity through a forest reserve network is a tradeoff between distributing risk over many reservesrepresenting the full spectrum of environmental settings (the goal of this analysis stage) and minimizingthe probability of loss in each individual reserve (the goal of the size criteria analysis stage).In order to identify a representative network of matrix forest blocks, a dataset was developed to assessthe geophysical character of the landscape, which can also be used to approximate the distribution andcomposition of many community assemblages across those landscapes. The close relationship of thephysical environment to ecological process and biotic distributions underpins the ecological sciences,and given the changing climate and associated changing species distributions, geophysical diversity maybe an appropriate target for conservation planning (Anderson and Ferree 2010) Research has repeatedlydemonstrated especially strong links between ecosystem pattern and processes, climate, bedrock, soils,and topography.To quantify the physical diversity of the Southern Blue Ridge, we developed individual 30m cell GISdatasets of elevation, geology, and landforms, and integrated them into a single unit called theEcological Land Unit (ELU). Each ELU code represents a specific landform (e.g., cove), composed of aspecific bedrock type (e.g., calcareous), within a specific elevation zone (e.g., low elevation). The ELUdataset describes only the physical information, but it can be combined with land use or land covermaps to approximate vegetation types. We present a brief description of each of the ELU dataset layersbelow.ELEVATIONElevation is a strong predictor of the distribution of forest communities. Temperature, precipitation,and exposure commonly vary with changing altitude, as does the dominant vegetation type. Red sprucebegins to occur within northern hardwood forests at around 1,405m (4,500ft) and becomes moredominant with increasing elevation. Fraser fir species abundance increases above 1,720m (5,500ft) andcan occur in pure stands over 1,875m (6,000ft). Hemlock-dominated forests, along with white pine andhardwood mixed habitats, occur at middle elevations and often in small stands along north-facing slopesat elevations between about 762-1,220m (2,500-4,000ft). Variations of cove forest can be foundthroughout the Southern Appalachians at elevations from 305 to 1,372m (1,000-4,500ft). Appalachianoak forests composed of northern red, chestnut, white, and black oaks, in combination with many otherspecies (especially pignut and mockernut hickory and red maple) typically occur at elevations from 250mto 1,280m (820-4,200ft) elevations (Stephenson et al. 1993, Hunter et al. 1999).We used the following elevation cutoffs to map distinct elevation zones:Very HighHighMedium HighMedium LowLow5,500ft+4,200-5,500ft3,500-4,200ft2,300-3,500ft2,300ft or less13

GEOLOGYBedrock geology strongly influences area soil and water chemistry. Bedrock types also differ in howthey weather and in the physical characteristics of the residual soil type. Because of this, local lithologyis usually the principle determinant of soil chemistry, texture, and nutrient availability. Many ecologicalcommunity types are closely related to the chemistry and drainage of the soil or are associated withparticular bedrock exposures.We grouped bedrock units on the bedrock geology maps of Virginia, North Carolina, Tennessee, SouthCarolina, and Georgia into nine general classes, designed to have particular relevance to vegetationdistributions. The nine categories listed below were based on the chemical and physical properties ofthe soils derived from them, and are correlated with regional biodiversity patterns (Anderson and Ferree2010, Appendix C).Acidic sedimentary: Fine to coarse-grained, acidic sedimentary or meta-sedimentary rock. This groupincludes: mudstone, claystone, siltstone, non-fissile shale, sandstone, conglomerate, breccia, greywacke,and arenites. Metamorphic equivalents inlcude: slates, phyllites, pelites, schists, pelitic schists, andgranofels.Acidic shale: This group includes any fine-grained loosely compacted acidic fissile shale.Calcareous: Alkaline, soft, sedimentary or metasedimentary rock with high calcium content. This groupincludes: limestone, dolomite, dolostone, marble, and other carbonate-rich clastic rocks.Moderately Calcareous: Neutral to alkaline, moderately soft sedimentary or meta-sedimentary rockwith some calcium but less so than the calcareous rocks. This group includes: calcareous shales, pelitesand siltstones, calcareous sandstones, lightly metamorphosed calcareous pelites, quartzites, schists andphyllites, and calc-silicate granofels.Acidic Granitic: Quartz-rich, resistant acidic igneous and high grade meta-sedimentary rock. This groupincludes: granite, granodiorite, rhyolite, felsite, pegmatite, granitic gneiss, charnockites, migmatites,quartzose gneiss, quartzite, and quartz granofel.Mafic: Quartz-poor alkaline to slightly acidic rock. This group includes: (ultrabasic) anorthosite (basic),gabbro, diabase, basalt (intermediate), quartz-poor: diorite/ andesite, syenite/ trachyte, greenstone,amphibolite, epidiorite, granulite, bostonite, and essexite.Ultramafic: Magnesium-rich alkaline rock. This group includes: serpentine, soapstone, pyroxenites,dunites, peridotites, and talc schist.Coarse Surficial Sediment: This group includes deep unconsolidated sand and gravel.Fine Surficial Sediment: This group includes deep unconsolidated silt and mud.14

LANDFORMLandforms anchor and control terrestrial ecosystems by breaking broad landscapes into localtopographic units and creating meso- and micro-climates. Topography is largely responsible for localvariation in solar radiation, soil development, moisture availability, and susceptibility to wind and otherdisturbances. To map SBR ecoregional landforms, we created an 11-part GIS landform model thatdelineated local environments, representing distinct combinations of moisture, radiant energy,deposition, and erosion. The model, based on land surface models developed for soil mapping,categorizes various combinations of slope, land position, aspect, and moisture accumulation.Development methods were based on Fels and Matson (1997) and are described in detail in Anderson etal. 2012. The model is derived from a 30-meter Digital Elevation Model and maps the followinglandforms (Figure 4):1) Cliff or steep slope (includes cliffs and steep slopes of warm and cool aspects)2) Summit or ridgetop (includes flat summit, upper ridges, and slope crests)3) Northeast sideslope (includes moderately steep sideslopes of cooler aspects)4) Southwest sideslope (includes moderately steep sideslopes of warmer aspects)5) Cove or slope bottom (includes slope bottom flats, and coves of warm and cool aspects)6) Low hill7) Low hilltop flat8) Valley or toeslope9) Dry flat10) Wet flat11) Water (includes lakes, ponds,rivers and estuaries)ECOLOGICAL LAND UNITSTo integrate the elevation, geology, andlandform characteristics, we used GIS tocode every 30-meter cell in the regionwith its elevation zone, geology class,and landform type. For example, a cell at1,400 feet (elevation class 2000) ongranitic bedrock (substrate class 500) in awet flat (landform 31) would be coded2531 (Figure 4, Table 3). In total, theSouthern Blue Ridge Ecoregion had 453unique combinations of elevation zone,substrate type, and landform. TheseELU data were used to provide anabiotic characterization of each of thecandidate matrix forest blocks (Figure5).Figure 4. An example of landforms, used for Ecological Land Unitdevelopment, in Linville Gorge Wilderness Area in North Carolina.15

Table 3. Ecological Land Units Coding Scheme.Elevation class (ft) + Substrate class + Landform1000 (0-2300) 100 acidic sed/metased 4 steep slope2000 (2300-3500) 200 acidic shale 5 cliff3000 (3500-4200) 300 calc sed/metased 11 flat summit/ridgetop4000 (4200-5500) 400 mod. calc sed/metased 13 slope crest5000 (> 5500) 500 acidic granitic 21 Hilltop (flat)600 mafic/intermed granitic 22 Hill (gentle slope)700 ultramafic 23 NW-facing sideslope800 coarse sediments 24 SE-facing sideslope900 fine sediments 30 Dry flat31 Wet flat32 Valley/toe slope41 Flat at bottom of steep slope43 NW-facing cove/draw44 SE-facing cove/draw51 Polygonal rivers from NHD52 NHD lakes/ponds/reservoirsClassifying Matrix Forest Blocks by ELU ComponentsWe overlaid potential matrix forest block boundaries (meeting size and land cover criteria as describedabove) with the ELU data layer, and tabulated the area of each unique ELU code within each block,summarizing the types and amounts of physical features contained. We used this information to classifythe forest blocks, using a standard quantitative cluster analysis (CLUSTER and TWINSPAN) within PC-ORDversion 6 (McCune and Mefford 1992, McCune and Grace 2002) to aggregate the forest matrix blocksinto groups that shared a similar combination of physical features. This ordination process resulted ingroups of similar recognizable forest-landscape combinations, which we termed “ELU groups” (Figure 6).16

The analysis resulted in 25 ELU groups, based largely on differences in elevation zones and geologytypes. Among ELU groups, the types and proportions of landforms varied less (the largest variationswere related to the proportion of flats and wet flats in each block). We further subdivided some of thegroups into subgroups (labeled a, b, or c) based on both ecological characteristics and the geographicdistribution of the blocks. For example if a group contained two spatially-disjunct sets of blocks thatwere found in different physiographic regions, we split that ELU group into subgroups (Figure 6).The resulting ELU groups and subgroups, and the names of the matrix forest blocks they contain, arelisted below (arranged by elevation zone):LOW ELEVATION• Group 3: Low elevation acidic sedimentary blocks, mostly sideslopes with coves prevalent –Examples: Fort Mt., Hiwassee, Hiwassee Lake, Web Mt.• Group 4: Low elevation, mix of granite and acidic sedimentary, no summits, mostlysideslopes - Examples: Lower Chattooga, Mid-Chattooga, Sumter• Group 13b: Low elevation block on granite and mafic bedrock, steep landforms, valleys fewflats - Examples: South Mt. ConnectorLOW to MID ELEVATION• Group 1: Low to mid elevation, mixed acidic sedimentary and shale, mostly slopes -Examples: Cohutta, Lover's Leap, Sol Messer Mt., Meadow Creek Mt., Starr Mountain• Group 2: Low to mid elevation, acidic sedimentary with bits of granite or mafic, mostlysideslope - Examples: Amicalola West, Burnt/Sassafras, Dahlonega/Amicalola, Payne Mt.• Group 5: Low to mid elevation with acidic sedimentary, granite and mafic bedrock. Slopes,few flats - Examples: Bull Mountain, Fairy Stone, Pinnacles of Dan• Group 8: Low to mid elevation, very diverse geo: acidic sedimentary, granite, moderatelycalcareous and calcareous. Mostly steep features, few gentle hills - Examples: Max Patch• Group 13a: Low to mid elevation on moderately calcareous and granite bedrock, various landforms - Examples: Adams Mt., Globe, Linville Gorge• Group 14a: Low to mid elevation on granite and coarse surficial sediment. Sideslope, no wetflats - Examples: Poor Mountain• Group 14b Mid to low elevation on granite, Mostly slopes and slope crests, no wet flats -Examples: Mack’s Mountain• Group 14b Mid to low elevation on granite, Mostly slopes and slope crests, no wet flatsExamples: Mack’s Mountain• Group 14c: Mid to low elevation on granite, variety of landforms no wet flats Examples:Horse Trough Mt., Raven Cliffs, Unicoi, Blood Mt.• Group 15: Low to mid elevation on granite or with bits of mafic and mod calc, variouslandforms Examples: Honeycut Mt./Humpback Mt., Little Switzerland, Doughton/Stone Mt., ElkRidge, Rendezvous Mt., Vannoy, Yadkin Headwaters, Bent Mountain• Group 16-18: Low to mid elevation on granite with substantial component of acidicsedimentary, sideslopes and a wide variety of landforms Examples: Blue Wall, Green River,Chimney Rock, Great Smokies West, Hickory Nut Gorge, Jocassee, Mt. Bridge, Mt. Bridge West,Spring Creek Mt., Headwaters, Tiger, Upper Chattooga/Whitewater, Duncan Ridge, Rocky Mt.,Asheville Watershed, Grandfather Mt.17

MID ELEVATION• Group 6: Mid elevation block of calcareous limestone with steep slopes and sideslopes -Examples: Doe Mt.• Group 22-24: Mid elevation blocks on granite, coves prevalent along with sideslope, steepslopes - Examples: Big Bald, Roan Mt West, Cowee, Cullasaja, Newfound Mountains,Panthertown Valley, Pink Beds, Standing Indian, Looking Glass, Mt. Pisgah, Roy Taylor, RichlandBalsam, Shining Rock• Group 25: Mid elevation blocks with substantial mafic bedrock. Mostly sideslope - Examples:Amphibolite Mts., Roan Mt. East• Group 19b: Mid elevation block on granite, sideslopes and steep slope - Examples: Cathy'sCreek, N. Fork French Broad, Mount Rogers• Group 19a: Mid elevation on granite with substantial acidic sedimentary. Mostly sideslopeand valley - Examples: Dupont, Dupont West• Group 20b: Mostly mid- elevation and mostly granite and a variety of landforms - Examples:Big Laurel, Hurricane Mts., Pond Mt., Unaka , Mt. WilderMID to HIGH ELEVATION• Group 10: Mid to high elevation with moderately calcareous and acidic sedimentary. Mostlyslopes - Examples: Nantahala• Group 21: Mid to high elevation with limestone and granite, mostly sideslopes, and steepslopes - Examples: Jefferson/Iron Mt., Sugar GroveALL ELEVATION ZONES• Group 7: Diverse blocks with many elevation zones, and many types of sedimentarybedrocks. No wet flats, mostly sideslopes and steep slopes - Examples: Buffalo Mountain,Holsten Mt., Rocky Fork, Scioto-Stone Mt. Stone Mountain, The Snake• Group 9: All elevation zones, sedimentary bedrock with sideslopes, steep slopes and coves -Examples: Brasstown Bald, Cheoah, Cove Mt. WMA, Fires Creek, Great Smokies East, HarmonDen, Joyce Kilmer/Unicoi, Rich Mt., Yellow Creek• Group 11: All elevation zones on sedimentary and granite. Mostly sideslopes and steepslopes - Examples: Black Mts., Little Tennessee, Plott Balsams, Qualla, Rabun Bald• Group 12: All elevation zones, diverse blocks with granite, sedimentary and shale no flats orgentle slopes of any kind - Examples: Forge Mt. /Mt. Rogers, Nolichucky• Group 20a: All elevation zones and equal parts granite and sedimentary. Steep slopes, withfew flats or gentle slopes - Examples: Chunky Gal, Tri-County18

Figure 5. Southern Blue Ridge Potential Matrix Forest Blocks and Ecological Land Units.19

Figure 6. Southern Blue Ridge Early Matrix Forest Blocks by similar Ecological Land Unit Groupings.20

Figure 7. Prioritized Southern Blue Ridge Matrix Forest Blocks, with Tier 1 and 2 Priorities and Key Connectors Delineated.21

EVALUATING AND PRIORITIZING A NETWORK OF MATRIX FOREST BLOCKSThe final step of this process was to evaluate the matrix forest blocks and prioritize a subset forconservation focus. Following internal TNC ecoregion-wide staff meetings to delineate potential matrixforest blocks, expert meetings were set up in each of the five SBR states between October 2010 andFebruary 2011, to which a wide range of forest experts and partners were invited to attend and evaluatedraft matrix forest block results (Appendix A). State expert review meetings followed a process thatcompared and contrasted attributes of potential matrix forest blocks in each ELU group, first orientingexperts to block locations and basic properties, then examining compiled data and documentingcomments on the appropriateness of block boundaries, natural heritage and forest condition qualities,and land ownership patterns. Copious notes were taken during the expert meetings and can be found athttp://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspx.State expert teams evaluated detailed block information compiled for the process, as well as discussingpersonal knowledge of the areas. Additional information provided and/or discussed by participantsincluded:- The amount and type of conservation land within each block- The density of internal roads- The number and types of rare species and natural communities found within each block- The number and types of unique natural communities found in each block- The degree of development or agriculture within the block- Known old-growth sites (from Southern Appalachian Forest Coalition)- The type and number of unique ELUs within each block- Satellite or Air photo imagery from Google Earth- The degree of invasive species within each block- The condition of the forest within each block- Rare species and natural community information was provided by the Natural Heritageprograms from the five states and used with their permission. A summarized table of thisinformation can be found within Appendix E.Following expert meetings, TNC staff reviewed all compiled feedback and attributes and determinedconsensus around final block boundaries for each of the 107 potential matrix forest blocks. This reviewwas conducted systematically, by ELU-based groupings, with the goal of identifying at least one priorityblock within each of the 25 unique ELU groups in the SBR. This ensured that our final network ofprioritized matrix forest blocks would be not only be those relatively intact blocks of sufficient size, butalso representative of all geophysical variation in the region (Figure 7, Appendix D). Information aboutdecisions made at this meeting concerning final block boundaries and priorities can also be found athttp://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspx.22

RESULTS OF ANALYSISAll final SBR matrix forest blocks were categorized as Tier 1 (higher priority blocks), or Tier 2 (important,but lower conservation priority) (Figure 7, Appendix D). Some initial potential blocks were excludedfrom final delineated priority matrix forest block results for this analysis, but they may still merit otherconservation considerations. The team also decided to categorize nine blocks as “connector blocks 2 ;”those blocks that appear to have high connectivity value among and between blocks designated as Tier1 or Tier 2. The resulting GIS database of spatially-delineated Tier 1, Tier 2, and Connector SBR ForestMatrix Blocks and relevant metadata is available for download and use athttp://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspx.SIZE OF MATRIX FOREST BLOCKSOf the original 107 potential blocks, the final analysis resulted in the identification of 83 Southern BlueRidge priority matrix forest blocks (comprised of 50 Tier 1 and 24 Tier 2 blocks, totaling 4,721,577acres), plus nine connector blocks (Figure 7, Appendices D, E). The average size of all of the blocks was53,822 acres. Eighty percent of the final block acreage was categorized as Tier 1 priority (3,791,877acres). Amongst the Tier 1 matrix forest blocks, block size ranged from 19,338 acres (Doe Mountain) to340,182 acres (Great Smokies West), with the number of blocks falling into various size cohorts (Table4). Amongst the 929,700 acres of Tier 2 forest blocks, block size ranged from 13,733 (Chimney Rock) to78,235 acres (Lower Chattooga), with the number of blocks falling into various size cohorts (Table 5).Most of the unique SBR ELU groups were represented by at least one or two Tier 1 blocks, except forELU groups 10, 17a, and 17c (which are represented by Tier 2 blocks) and 15c and 19b (for which noblocks justified representation) (Tables 4, 5).Table 4. Tier 1 Matrix Forest Block Size Cohorts and ELU Groups Represented.Block Size Number ELU Groupings Represented in Cohortof Tier 1Blocks15,000 – 30,000 acres 6 5, 6, 7, 9, 14c30,001 – 50,000 acres 12 1, 4, 11, 12, 14a, 14b, 14c, 15a, 15b, 20b, 2450,001 – 70,000 acres 15 7, 14c, 16, 17b, 19a, 19c, 20a, 20b, 22a, 22b, 2570,001 – 100,000 acres 7 7, 11, 13a, 16, 18, 21, 22b100,001 – 150,000 acres 6 2, 3, 8, 9, 13a, 23150,001 acres or more 4 1, 9, 112 The designation of “connector blocks” was not made in a concertedly consistent fashion during the reviewprocess. It is expected that a GIS connectivity analysis will be conducted for the SBR ecoregion that will refinethese designations and likely point to additional areas of high connectivity potential.23

Table 5. Tier 2 Matrix Forest Block Size Cohorts and ELU Groups Represented.Block SizeNumberof Tier 1Blocks10,000 – 30,000 acres 8 1, 2, 3, 9, 10, 15a, 16ELU Groupings Represented in Cohort30,001 – 50,000 acres 12 4, 11, 12, 13a, 15b, 16, 17a, 17b, 17c, 20b, 22a, 22b50,001 – 70,000 acres 2 9, 20a70,001 – 100,000 acres 2 4, 11100,001 – 150,000 acres 0 n/a150,001 acres or more 0 n/aCONSERVATION STATUS AND CONDITION OF MATRIX FOREST BLOCKSGAP Status as a Proxy for Protection and ManagementComputed characteristics, including land ownership and conservation status, are summarized by blockand ELU grouping in Appendix E. One way to assess current condition or conservation status in relationto protection and management intentions of these blocks is to use GAP Status designations (Crist et al.1998). Using the GAP Status, we determined that a large percentage of the blocks are already withinsome type of conservation status, with land either secured explicitly for the conservation of biodiversityand ecological processes (e.g., GAP 1 and 2 Status), or land secured for multiple uses (e.g., GAP 3 Status,Figure 8, Crist et al. 1998). We recognize that GAP Status designations are not the only way to assessthe current level of protection or intensity of intended management activities and can sometimes bemisinterpreted 3 , but we have used these designations to coarsely assess our blocks on a landscapescale.Additional work with land managers and stakeholders is necessary to better determine currentconservation status, management intentions, and opportunities for spatially-explicit ecologicallycompatiblemanagement goals, using these results.On average, 7% of each block was designated as GAP 1 Status (or 7, 411 acres), with an average of 3%(1,571 acres) and 40% (21,675 acres) of each block designated as GAP 2 or 3 Status, respectively (Figure8). Half of the area within all of the matrix forest blocks is unsecured (Appendix E). The majority of landwithin GAP 1 Status is owned by the National Park Service (Great Smoky Mountains National Park) andthe US Forest Service (designated Wilderness Areas within National Forests). These lands are consideredpermanently protected and are usually managed for natural or mimicked (through managementactivities) disturbance events. The majority of land within GAP 2 Status is owned by State WildlifeAgencies and State Parks. These lands also are considered permanently protected, but they aredesignated to receive use or management which may result in some degradation of the quality ofexisting natural communities. The majority of land within GAP 3 Status is owned by US Forest Service(multiple use areas within the National Forests). These lands are considered to have some permanent3 E.g., GAP 1 designated areas can also be considered ecologically-degraded (e.g., as a result of the inability toeffectively reintroduce ecologically-appropriate fire into a system).24

protection, but are subject to natural resource extractive uses that can result in more significantdegradation of biodiversity or natural ecological integrity.Fragmenting Features and Nonforest Land Cover Within BlocksWe determined the size of the largest roadless (unfragmented) area within each of the blocks. Fifty-fivepercent of all of the blocks (26,774 acres) were determined to be roadless, while the average roaddensity per block was 3.45 miles/1,000 acres.Nonforest land cover information within matrix forest blocks was also analyzed and summarized(Appendix E). On average, pasturelands and developed open space land cover classes each comprised3% of blocks. An average of 1% was classed as evergreen plantations or managed pines, and 1% wasclassed as successional shrub/scrub land within blocks.Species and Natural CommunitiesStatistics regarding known occurrences of tracked species and natural communities were also calculated(Appendix E). On average, for each block, there were:• 12.1 tracked natural community occurrences,• 62 tracked vascular plant species occurrences,• 36.6 tracked fish species occurrences,• 25.6 tracked amphibian species occurrences,• 11.8 tracked non-vascular plant species occurrences,• 10.9 tracked mammal species occurrences,• 8.2 tracked invertebrate species occurrences,• 2.9 tracked bird species occurrences,• 1.9 tracked reptile species occurrences,There were 172 total known element occurrences overlapping matrix forest blocks, representing a totalof 41 species (total taxa) (Appendix E). The majority of the Conservancy’s portfolio sites and rare speciesoccurrences that were identified in the SBR Ecoregional Assessment in 2000 are nested within thesematrix forest blocks (TNC & SAFC 2000). Further analysis is needed to determine how best toincorporate those originally identified portfolio sites (i.e., islands of biodiversity) not nested within thisnewly identified contiguous, connected matrix forest habitat (Ward et al. 2011).25

Figure 8. SBR Matrix Forest Blocks Overlapping GAP Status (a proxy for determining the level of protection and intensity of management expected).26

CONCLUSIONSThe goal of this analysis was to identify and evaluate a network of representative contiguous forests inthe Southern Blue Ridge, which are large enough to sustain forest ecosystem processes and theembedded conservation targets within them. The Nature Conservancy’s vision for the conservation ofthese identified priority SBR matrix forest blocks is to work with partners to: (1) ensure adequate longtermprotection and ecologically-compatible management of this network of representative matrixforest blocks, with embedded core forests surrounded with multiple working forest buffers (Noss andCooperrider 1994, Figure 2), (2) retain connectedness of forest cover among and between those matrixforests and across the landscape, and 3) abate region-wide forest threats.Overall, the Southern Blue Ridge matrix forest blocks that were delineated, screened, classified,evaluated, and prioritized have resulted in considerable information that can be used to informconservation strategies in this region. These results close a significant gap in our understanding of thespatial context and characteristics of large contiguous forests in the Southern Blue Ridge, which areimportant for the people inhabiting the region, as well as for the fundamental ecosystem services thatthey provide. The Conservancy believes the results of this analysis will provide a clearer direction fornear-term conservation priorities, and foster a more focused set of conservation actions around whichTNC and partners can organize and cooperate.Furthermore, we recommend that TNC operating units and interested partners within the SBR, bothindividually and jointly, use the priority matrix forest block information derived from this exercise toenhance our understanding of the highest conservation priorities and strategies in the SBR. Examples ofthis include:• Consideration of new or redrawn priority conservation landscapes;• Assessment of collaborative partnership opportunities (e.g., major land management entitieswithin priority blocks, with whom collaboration could improve forest condition in alignment withmatrix forest, core forest, and working forest buffer principles (see Figure 2);• Reassessment of land protection opportunities that could lead to improved forest blockmanagement practices in priority areas, in alignment with matrix forest, core forest, and workingforest buffer principles; and• Reassessment of the appropriate balance and strategic success of both localized and landscapescaleland protection, land management, and policy-oriented strategies, coordinated andexecuted across multiple (versus individual) program offices and operating units.It is important to note that our regional forest network model is based on the concept that in an “ideal”situation, the full spatial extent of all priority matrix forests would be managed primarily for biodiversityprotection, long term ecological monitoring, and low impact human use (with the overall goals ofmaintaining forest cover, habitat diversity, rare species, significant natural communities, and waterquality), we recognize that this is not practical. With this in mind, we recommend that within eachmatrix forest block, single or multiple core forests of adequate size be identified and managed primarilyfor biodiversity and the maintenance of natural processes, while surrounding multiple use buffer areasare managed for a wider variety of biological, social, and economic values (Figure 2). Additionally, it isworth mentioning that matrix forest blocks and their associated core forests will require varying levelsof active ecological management (though certainly there will be areas within these biological cores that27

are essentially unmanaged) to achieve the desired future conditions of these important forests. It is alsocritical to emphasize that the identified core forests can support a variety of low impact recreation.Therefore, we conducted a draft analysis to spatially-identify the best opportunities for these coreforests (i.e., we delineated unfragmented blocks (≥5,000 acres) of heterogeneous, interior forestlandscapes that, being nested within the matrix forest blocks, provide the opportunity for relativelynatural processes to occur or be mimicked through management, resulting in a healthy range ofstructural and compositional forest attributes). These delineated draft interior core forests are intendedto be used as an initial discussion point with partnering agencies (e.g., U.S. Forest Service) and are by nomeans definitive in their extent. The results of these delineations are dependent upon the inputs ofstatewide and national datasets, and are therefore best suited for regional, broad-scale planning. Finerscaleplanning should incorporate more locally-based county and city datasets and air-or- groundtruthingto reflect the most current conditions and appropriate boundaries of these delineated interiorcore forests. The methods and results of this 2012 SBR Core Forest Analysis will be summarized in areport, expected to be released in May 2013.In addition to the identification of core forests, TNC Southern Blue Ridge staff has identified additionalpriority follow-up steps to this SBR Matrix Forest Block Analysis, including conducting Climate ChangeResiliency Analyses related to connectivity of terrestrial and freshwater landscapes in the southeasternUnited States. TNC is currently undertaking both Southeastern Terrestrial and Freshwater ResilienceAnalyses, to complement completed analyses for the northeastern US. The southeastern US ResiliencyAnalyses are expected to be complete by 2014 (for terrestrial) and 2013 (for freshwater). Once theseanalyses are complete, SBR TNC staff and partners should revisit the results of the Matrix Forest andCore Forest Analyses in order to better incorporate areas of expected high resilience in the face ofclimate change and redefine terrestrial and freshwater priorities accordingly. Furthermore, futureconsideration of Ecological Land Units, classified and grouped based on watershed-scale is necessary tobetter incorporate and prioritize representative freshwater systems.Photo by John Warner28

LITERATURE CITEDAnderson, M.G. 2008. Conserving Forest Ecosystems: Guidelines for Size, Condition and LandscapeRequirements, in R. A. Askins, G. D. Dreyer, G. R. Visgilio and D. M. Whitelaw (eds.), Saving BiologicalDiversity: Balancing Protection of Endangered Species and Ecosystems, pp. 119-136. Springer-Verlag,New York, New York.Anderson, M.G., M. Clark, and A. Olivero Sheldon. 2012. Resilient Sites for Terrestrial Conservation inthe Northeast and Mid-Atlantic Region. The Nature Conservancy, Eastern Conservation Science.168pp. http://conserveonline.org/workspaces/ecs/documents/resilient-sites-for-terrestrialconservation-1Anderson, M. G., and C. Ferree. 2010. Conserving the Stage: climate change and the geophysicalunderpinnings of species diversity. PLoS ONE 5(7):E11554.DOI:10.1371/journal.pone.0011554.Anderson, M.G., S. Bernstein, F. Lowenstein, N. Smith, and S. Pickering. 2004. Determining the size ofeastern forest reserves. The Nature Conservancy, Eastern Region. Boston, MA.Barden, L. S., and F. W. Woods. 1974. Characteristics of lightning fires in the southern Appalachianforests. p. 345-361. In: Proceedings, 13th Tall Timbers fire ecology conference; 1973 October 14-15;Tall Timbers Research Station. Tallahassee, FL.Crist, P.J., B. Thompson, T. C. Edwards, C. G. Homer, S. D. Bassett. 1998. Mapping and Categorizing LandStewardship. A Handbook for Conducting Gap Analysishttp://www.gap.uidaho.edu/handbook/Stewardship/default.htmCroy, Steve. 2011. Personal communication.Dudley, N. and S. Stolton. 2003. Running Pure: The importance of forest protected areas to drinkingwater. World Bank/WWF Alliance for Forest Conservation and Sustainable Use Report. Available at:http://www.forestsforwatersheds.org/storage/runningpurereport.pdfEhrlich, P.R., and D.D.Murphy, 1987. Conservation lessons from long-term studies of the checkerspotbutterflies. Conservation Biology 1:122-131.ESRI 2009. ESRI Data and Maps U.S. and Canada Streets Cartographic. ESRI Geographic Data Technology,Inc., Redlands, CA.Fahrig, L. 2003 Effects of habitat fragmentation on biodiversity. Annu. Rev. Ecol. Evol. Syst. 2003.34:487–515.Fels, J. E. and K. C. Matson. 1996. A cognitively based approach for hydro-geomorphic land classificationusing digital terrain models. In: 3rd International Conference on Integrating GIS and EnvironmentalModeling. National Centre for Geographic Information and Analysis, Santa Barbara, California. CD-ROM. www.ncgia.ucsb.edu/conf/SANTA_FE_CD-ROM/sf_papers/fels_john/fels_and_matson.html29

Fesenmyer, K.A. and N. L. Christensen, JR. 2010, Reconstructing Holocene fire history in a southernAppalachian forest using soil charcoal. Ecology, 91(3), 2010, pp. 662–670.Forman, R.T.T. 1995. Land mosaics: the ecology of landscapes and regions. Cambridge University Press.Cambridge, 632 p.Franklin, J.F. 1993. Preserving biodiversity: species, ecosystems, or landscapes? Ecological Applications3: 202-205.Fry, J., Xian, G., Jin, S., Dewitz, J., Homer, C., Yang, L., Barnes, C., Herold, N., and Wickham, J.2011. Completion of the 2006 National Land Cover Database for the Conterminous UnitedStates, PE&RS, Vol. 77(9):858-864.Greenberg, C.H., and W.H. McNab. 1998. Forest disturbance in hurricane-related downbursts in theAppalachian mountains of North Carolina. Forest Ecology Management. 104: 179-191.Greenberg, C. H., B. S. Collins, and F. R. Thompson III. 2011. Sustaining young forest communities:Ecology and management of early successional habitats in the central hardwood region, USA.Managing Forest Ecosystems. Volume 21. 308 p.Gunderson, L. H. 2000. Ecological resilience--in theory and application. Annual Review of Ecology andSystematics 31:425-439.Hamel, P. B. 1992. Land manager’s guide to birds of the South. Gen. Technical Report SE-22. Asheville,NC. US Department of Agriculture Forest Service, Southeastern Forest Experiment Station. 437 p.Hartley, M.J., & M.L. Hunter, Jr. 1997. A meta-analysis of forest cover, edge effects, and artificial nestpredation rates. Conservation Biology, 12(2), 465-469.Hunter, M.L., Jr. 1996. Fundamentals of Conservation Biology. Blackwell Science. Cambridge.Massachusetts, 482 p.Hunter, C., Katz, R., Pashley, D. and B. Ford. 1999. Partners in Flight Bird Conservation Plan for TheSouthern Blue Ridge (Physiographic Area 23) Version 1.0. Partners in Flighthttp://www.partnersinflight.org/bcps/pl_23sum.htmHunter, W. C., D.A. Buehler, R. A. Canterbury, J. L. Confer, and P. B. Hamel. 2001. Conservation ofdisturbance-dependent birds in eastern North America. Wildlife Society Bulletin. 29 (2): 440-455.Keys, Jr., J., C. Carpenter, S. Hooks, F. Koenig, W.H. McNab, W. Russell, and M.L. Smith. 1995. Ecologicalunits of the eastern United States – first approximation (CD-Rom), Atlanta, GA: U.S. Department ofAgriculture, Forest Service. http://www.srs.fs.usda.gov/econ/data/keys/index.htm.30

Lafon, C.W., Hoss, J.A. and Grissino-Mayer, H.D. 2005, The contemporary fire regime of the CentralAppalachian Mountains and its relation to climate. Physical Geography 26(2) pp. 126-146.Lande, R. (1988). Genetics and demography in biological conservation. Science 241: 1455–1460.Loftis, David. 2011. Personal communication.Martin, T.E., and Finch, D.M. (Eds.) 1995. Ecology and management of neotropical migratory birds: asynthesis and review of critical issues. New York: Oxford Press.McCune, B. and M.J. Mefford. 1999. PC-ORD for windows (software). Multivariate analysis of ecologicaldata, Version 4.XX. MjM Software, Gleneden Beach, OR, USAMcCune, B. and J. B Grace. 2002. Analysis of Ecological Communities. MjM software design. Oregon USA.300 pp. www.pcord.com.McIntyre, N.E. 1995. Effects of forest patch size on avian diversity. Landscape Ecology 10(2):85-99.Noss, R. F. and A. Y. Cooperrider. 1994. Saving nature’s legacy: protecting and restoring biodiversity.Washington, D.C.: Island Press.Paton, P.W.C. (1994). The effect of edge on avian nest success: how strong is the evidence?Conservation Biology 8:17–26.Pickett, S.T.A., and J.N. Thompson. 1978. Patch dynamics and the size of nature reserves. BiologicalConservation 13:27-37.Poiani, K.A., B.D. Richter, M. G. Anderson, and H. E. Richter. 2000. Biodiversity conservation at multiplescales. BioScience 50(2):133-146.Poole, A., and Gill, F. (Eds.) (1992-ongoing). Birds of North America No 1–600. Philadelphia: TheAmerican Ornithologist’s Union; Washington, DC: The Academy of Natural Sciences.Ray, David. 2013. Personal communication.Robbins, C.S., Dawson, D.K., Dowell, B.A. 1989. Habitat area requirements of breeding forest birds of theMiddle Atlantic states. Wildlife Monographs 103:1–34.Runkle, J.R. 1981. Gap regeneration in some old growth forests of the eastern United States. Ecology62:1041-1051.Runkle, J.R. 1982. Patterns of disturbance in some old-growth mesic forests of eastern North America.Ecology 63:1533-1546.31

Stephenson, S. L., A. N. Ash, and D. F. Stauffer. 1993. Appalachian oak forests. Pages 255-304 in W. H.Martin, S. G. Boyce, and A. C. Echternacht, eds., Biodiversity of the Southeastern United States,Upland Terrestrial Communities. John Wiley and Sons, Inc. New York. 373 pages.The Nature Conservancy and Southern Appalachian Forest Coalition. 2000. Southern Blue RidgeEcoregional Conservation Plan: Summary and Implementation Document. The Nature Conservancy:Durham, North Carolina.U.S. Census Bureau. 2010. 2010 Census of Population and Housing, Demographic Proile Summary File:Technical Documentation, 2011.U.S. Environmental Protection Agency (USEPA) and the U.S. Geological Survey (USGS). 2008. NationalHydrography Dataset Plus – NHDPlus. Digital dataset. Accessed online at: http://nhd.usgs.gov/.Ventyx Corporation. 2010. Velocity Suite: EV Energy Map. Accessed online in late 2010 at:http://www.ventyx.com/velocity/ev-energy-map.asp.Ward, J., Agostini, V., Anderson, M., Burns, C., Doran, P., Fargione, J., Groves, C., Hanners, L., Hoekstra,J., Marshall, R., Morrison, S., Palmer, S., Shaw, D., and J. Smith. 2011. Stepping up to the Challenge: AConcept Paper on Whole System Conservation. The Nature Conservancy. 6 pp.32

Appendix A: State Expert Review Teams and TNC Analysis Team MembersState Name Agency DateE.Div. Mark Anderson TNC select meetingsE.Div. John Prince TNC all meetingsGA Nathan Klaus GA Dept. of Natural Resources November 12, 2010GA Tom Patrick GA Dept. of Natural Resources November 12, 2010GA Jon Ambrose GA Dept. of Natural Resources November 12, 2010GA Jason Wiesnewski GA Dept. of Natural Resources November 12, 2010GA Tom Govus NatureServe contractor November 12, 2010GA Malcolm Hodges TNC November 12, 2010GA Randy Tate TNC November 12, 2010GA Wade Harrison TNC November 12, 2010GA Sara Gottlieb TNC November 12, 2010GA Deron Davis TNC November 12, 2010GA Jim Wentworth USFS November 12, 2010GA Joanne Baggs USFS November 12, 2010NC Curtis Smalling Audubon October 27, 2010NC Kevin Caldwell independent biologist October 27, 2010NC Dennis Herman NC Dept. of Transportation October 27, 2010NC Marshall Ellis NC Division of Parks and Recreation October 27, 2010NC Mike Schafale NC Natural Heritage Program October 27, 2010NC Steve Hall NC Natural Heritage Program October 27, 2010NC James Padgett NC Natural Heritage Program October 27, 2010NC Ed Schwartzman NC Natural Heritage Program October 27, 2010NC Kendrick Weeks NC Wildlife Resources Commission October 27, 2010NC Lori Williams NC Wildlife Resources Commission October 27, 2010NC Christine Kelly NC Wildlife Resources Commission October 27, 2010NC Jay Leutze Southern Appalachian Highlands Conservancy October 27, 2010NC David Ray TNC all meetingsNC Rick Studenmund TNC October 27, 2010NC Merrill Lynch TNC October 27, 2010NC Megan Sutton TNC October 27, 2010NC Catherine Burns TNC October 27, 2010NC Carolyn Wells U.S. Fish and Wildlife Service October 27, 2010NC Gary Kauffman USFS, National Forests in NC October 27, 2010SC Patrick McMillan Clemson University October 26, 201033

State Name Agency DateSC Vic Shelburn Clemson University October 26, 2010SC Rob Baldwin Clemson University October 26, 2010SC Ed Pivorun Clemson University October 26, 2010SC Joe Pollard Furman University October 26, 2010SC Mark Hall SC Dept. of Natural Resources October 26, 2010SC Stan Hutto SC Dept. of Parks, Recreation, and Tourism October 26, 2010SC Tim Lee SC Dept. of Parks, Recreation, and Tourism October 26, 2010SC Kristen Austin TNC October 26, 2010SC Sarah Fraser TNC October 26, 2010SC Eric Krueger TNC October 26, 2010SC Colette DeGarady TNC October 26, 2010SC Maria Whitehead TNC October 26, 2010SC Jeff Magniez USFS October 26, 2010SC Tom Waldrop USFS Research Station-Clemson October 26, 2010TN Jamey Donaldson independent biologist February 18, 2011TN Jesse Webster NPS, GSMNP February 18, 2011TN Hugh Irwin Southern Appalachian Forest Coalition February 18, 2011TN Pete Wyatt TN Wildlife Resources Agency February 18, 2011TN Alex Wyss TNC February 18, 2011TN Sally Palmer TNC February 18, 2011TN Katherine Medlock TNC February 18, 2011TN Geoff Call U.S. Fish and Wildlife Service February 18, 2011TN Ed Clebsch University of TN-Knoxville February 18, 2011TN Mark Pistrang USFS, Cherokee National Forest February 18, 2011TN Joe McGuiness USFS, Cherokee National Forest February 18, 2011TN Josh Kelly Wild Law February 18, 2011VA Angela Watland TNC November 4, 2010VA Claiborne Woodall VA Dept. of Conservation and Recreation November 4, 2010VA Steve Croy USFS November 4, 2010VA Allen Boynton VA Dept. of Game and Inland Fisheries November 4, 2010VA Joe Miller VA Forestry November 4, 201034

Appendix B: Fragmenting Features: Detailed GIS infoRoads: Roads were considered the primary feature that fragments forested landscapes in the Southern Blue Ridge.Detailed road GIS data (ESRI 2009) were evaluated against aerial imagery (Google Earth and ESRI ArcGIS Onlineaccessed 2010-2011) for accuracy and completeness. This GIS vector dataset classed roads from most intensivelytraveled (i.e., Interstates) to least intensively traveled roads (including some driveways and trails). While thegeographic extent and detail of road lines were good, road classifications were not always consistently applied,especially for middle-sized and smaller road classes. Nonetheless, the group settled on using road classes 1through 4 which roughly equates to Interstates (1), U.S. Routes and some State Routes (2), State Routes and oldhighways (3) and exit and on ramps, service roads, and rest area roads (4), were determined to be fragmenters.Power Transmission Lines and Railroads: The team acquired a vector GIS dataset from Ventyx (January 2010),containing power transmission line and railroad location information, and assessed its accuracy via visualcomparisons with recent aerial imagery (Google Earth). The team noted large variations in the width of vegetationcuts made for power lines, as well as in how intensively these areas were managed to remain non-forested. Inmany places (primarily in ravines and other depressions), no visible vegetation cuts were observed, as some powerlines were constructed well above the tree-line in these areas. Due to this variation, power transmission lines werenot considered a consistent fragmenting feature across the landscape, but instead were assessed for eachindividual block following draft delineation. The distance of power lines in each applicable potential block wascalculated and included in the Matrix Forest Block Report Statistics. The presence of railroads was also excluded asa consistent fragmenting feature in this landscape, due to the sporadic traffic, tunnels, and often narrow canopyopenings made for them, as well as their adjacency to roads or rivers. Similar to the power lines, the distance ofrailroads in each applicable potential block was calculated and included in block reports. These details areaccessibleathttp://www.conservationgateway.org/Files/Pages/SouthernBlueRidgeAnAnalysisofMatrixForests.aspx.Lakes and Rivers: The team used the National Hydrography Database (NHD Plus 2008) to assess water bodies(lakes) and rivers as potential fragmenting features. The team determined that all lakes and only large rivers wouldbe considered fragmenting features. Large rivers were determined to be those stream segments that drain over3,861 square miles. There were very few river segments that met this threshold in the ecoregion.Analysis Area (Southern Blue Ridge Ecoregional Boundaries): The Nature Conservancy’s Ecoregional Boundariesare derived from the U.S. Forest Service Ecoregions developed by Keys et al. (1995). The Southern Blue RidgeEcoregion “analysis area” was considered the ecoregional boundaries, buffered (extended out) by five miles. Alldata was clipped and assessed using this five-mile ecoregional buffered polygon in order to avoid excludingrelevant, connected forested roadless areas and other data of interest that do not follow ecoregional lines.Developed/Disturbed Areas: Developed and otherwise disturbed areas were not used as fragmenting features todelineate the GIS-generated iteration of matrix forest block boundaries. Road data was a good proxy for excludingdeveloped areas from this analysis because most roadless areas in developed landscapes were below the minimumacreage threshold for matrix forest block consideration. Urban and other built environments were visuallyassessed, using the most recent aerial imagery available, and used to edit/ refine draft matrix forest polygons inpartner and other team meetings later in the process.35

Appendix C: Southern Blue Ridge Geology ClassesGeology class Lithotypes Metaequivalents100: ACIDICMudstone,(Low grade:)SEDIMENTARY /claystone, siltstone, slates, phyllites,METASEDIMENTARY: fine- non-fissile shale, pelites; (Modto coarse-grained, acidic sandstone,grade:) schists,sed/metased rock conglomerate, pelitic schists,breccia, greywacke, granofelsarenites200: ACIDIC SHALE: Finegrainedacidic sedimentaryFissile shalesrock with fissile textureCommentsLow to moderatelyresistant rocks typical ofvalleys and lowlands withsubdued topography; puresandstone and metasedimentsare moreresistant and may form lowto moderate hills or ridgesLow resistance; producesunstable slopes of fine talus300: CALCAREOUSSEDIMENTARY / META-SEDIMENTARY:basic/alkaline, softsed/metased rock withhigh calcium content400: MODERATELYCALCAREOUSSEDIMENTARY /METASED: Neutral tobasic, moderately softsed/metased rock withsome calcium but less sothan aboveLimestone, dolomite,dolostone, othercarbonate-richclastic rocksCalc shales, calcpelites andsiltstones, calcsandstonesMarbleLightly to mod.metamorphosedcalc pelites andquartzites, calcschists andphyllites, calcsilicategranofelsLowlands and depressions,stream/river channels,ponds/lakes, groundwaterdischarge areas; soils arethin alkaline clays, highcalcium, low potassium;rock is very susceptible tochemical weathering; oftenunderlies prime agriculturalareasVariable group dependingon lithology but generallysusceptible to chemicalweathering; soft shalesoften underlie agriculturalareas36

Geology class Lithotypes MetaequivalentsComments500: ACIDIC GRANITIC:Quartz-rich, resistantacidic igneous and highgrade meta-sedimentaryrock; weathers to thincoarse soilsGranite,granodiorite,rhyolite, felsite,pegmatiteGranitic gneiss,charnockites,migmatites,quartzosegneiss,quartzite, quartzgranofelsResistant, quartz-rich rock,underlies mts and poorlydrained depressions;uplands & highlands mayhave little internal reliefand steep slopes alongborders; generally sandynutrient-poor soils600: MAFIC /(Ultrabasic:)INTERMEDIATE GRANITIC: anorthositequartz-poor alkaline to (Basic:) gabbro,slightly acidic rock, diabase, basaltweathers to clays (Intermediate,quartz-poor:)diorite/ andesite,syenite/ trachyte700: ULTRAMAFIC:magnesium-rich alkalinerockGreenstone,amphibolites,epidiorite,granulite,bostonite,essexiteSerpentine, soapstone, pyroxenites,dunites, peridotites, talc schistsModerately resistant; thin,rocky, clay soils, sl acidic tosl basic, high in magnesium,low in potassium; moderatehills or rolling topography,uplands and lowlands,depending on adjacentlithologies; quartz- poorplutonic rocks weather tothin clay soils withtopographic expressionsmore like graniteThin rocky iron-rich soilsmay be toxic to manyspecies, high magnesium tocalcium ratios often containendemic flora favoring highmagnesium, low potassium,alkaline soils; upland hills,knobs or ridges37

Appendix D: Matrix Forest Block InformationTier BlockName OrdGroupLabelAcresconnector Bent Mt. 15c 4 31925.5connector Cowee 20a 16 63642.1connector Elk Ridge 15b 22 51881.8connector Glade Mt. 21 25 32183.7connector Rendezvous Mt. 15b 63 41992.1connector Scioto-Stone Mt. 7 69 15440.7connector South Mt. Connector 13b 71 11265.6connector Stone Mt. 7 74 24352.4connector Vannoy 15b 81 45355.4T1 Amphibolite Mts. 25 1 51023.2T1 Asheville18 2 72746.4Watershed/Curtis CreekT1 Bald Mts./Rocky Fork 7 3 97470.9T1 Black Mts. 11 7 71337.8T1 Blood Mt. 14c 8 20206.3T1 Blue Wall 16 9 56339.7T1 Brasstown Bald 9 10 23504.7T1 Buffalo Mt. 7 11 21424.0T1 Bull Mt. 5 12 21803.9T1 Chunky Gal 20a 14 51993.4T1 Cohutta 1 15 195868.2T1 Dahlonega/Amicalola 2 17 115582.0T1 Doe Mt. 6 18 19337.6T1 Doughton/Stone Mt. 15b 19 38053.0T1 Dupont/Mt. Bridge 19a 21 53812.5T1Grandfather Mt./AdamsMt./John's River13a 27 109890.1T1 Great Smokies East 11 28 254767.9T1 Great Smokies West 9 29 340182.5T1 Harmon Den/Max8 31 113580.3Patch/Hurricane Mts.T1 Headwaters 17b 32 52211.9T1 Hickory Nut Gorge 16 33 65899.238

Tier BlockName OrdGroupLabelAcresT1 Hiwassee 3 34 100197.3T1 Holsten Mt. 7 35 58984.2T1 Horsetrough Mt. 14c 37 31871.9T1 Iron Mt. East 19c 38 69166.4T1 Jocassee 16 39 82215.6T1 Joyce Kilmer/Unicoi9 40 148829.8Mts./Slick Rock NorthT1 Joyce Kilmer/Unicoi9 41 230016.7Mts./Slick Rock SouthT1 Linville Gorge 13a 42 72650.5T1 Little Switzerland 15a 43 30207.8T1 Looking Glass/Roy23 45 139871.7Taylor/Cathy's CreekT1 Lover's Leap 1 46 38842.0T1 Macks Mt. 14b 48 42005.6T1 Mid-Chattooga 4 50 38681.0T1 Mt. Rogers 21 52 75884.1T1 Nolichucky 12 54 33877.0T1 Pink Beds 22b 57 57300.1T1 Pond Mt. 20b 59 69800.2T1 Poor Mt. 14a 60 45353.1T1 Rabun Bald 11 61 49823.7T1 Raven Cliffs14c 62 27077.3WildernessT1 Richland Balsam 24 65 30703.7T1 Roan Mt West 22a 66 55225.0T1 Roan Mt. East 25 67 65580.8T1 Shining Rock 24 70 35818.2T1 Standing Indian 22b 73 95508.2T1 The Snake 7 76 55846.4T1 Unaka Mt. Wilderness 20b 78 46593.9T1 Unicoi 14c 79 58029.1T1 Upper17b 80 58880.2Chattooga/Three StateT2 Big Bald 22a 5 32524.5T2 Big Laurel 20b 6 49652.7T2 Chimney Rock 16 13 13733.239

Tier BlockName OrdGroupLabelAcresT2 Duncan Ridge 17c 20 49468.7T2 Fires Creek 9 23 60413.3T2 Forge Mt./Rogers12 24 37112.3RidgeT2 Globe 13a 26 35589.1T2 Green River 16 30 30218.8T2 Honeycut15a 36 15930.1Mt./Humpback Mt.T2 Little Tennessee 11 44 74459.7T2 Lower Chattooga 4 47 78235.5T2 Meadow Creek Mt./1 49 17612.8Chuckey Mt.T2 Mt. Bridge West 16 51 14065.6T2 Nantahala 10 53 16621.3T2 Panthertown Valley 22b 55 49657.6T2 Payne Mt. 2 56 15275.1T2 Plott Balsams 11 58 48520.4T2 Rich Mt. 9 64 29728.2T2 Rock Creek/Tails3 68 27047.9Creek/Cold SpringT2 Spring Creek Mt. 17a 72 40589.6T2 Sumter 4 75 34910.8T2 Tri-County 20a 77 68618.6T2 White Water 17b 82 39868.5T2 Yadkin Headwaters 15b 83 49846.040

Appendix E: Final Southern Blue Ridge Matrix Forest Block Statistics (*see Appendix F for key to column headers)ELU_GroupBlock Name Revised Name States AcresGA,NC, 195,861 Cohutta CohuttaTN81 Lover's Leap Lover's Leap NC,TN 38,8421 Sol Messer Mt. Sol Messer Mt. TN 39,3121Average 91,341Dahlonega/Amicalo Dahlonega/Amical102,222 la?olaGA9Burnt/Sassafras2 Mt.?Largest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 340,752 2145,550126,975 23 0 65 12 30299 2720,849 54 28,720 0 0 74 26 55 74 611,11143 28 16,803 0 0 43 57 11 024,23 15,18178 34 3 57,499 8 0 61 32 32 1 1136,3624,64196 364 54,494 0 24 53 23 93 9 9Burnt/SassafrasMt. GA 13,302 6,923 52 1,204 4,273 0 9 32 59 12 302 Amicalola West? Amicalola West GA 11,344 9,715 86 2,321 0 0 20 80 47 2614,59102 Payne Mt. Payne Mt. NC,TN 15,275 5 96 0 0 0 0 6 36216,90Average 35,538 0 68 6,462 15,272 0 8 26 65 39 73 2100,19 15,57233 Hiwassee Hiwassee TN7 3 16 4,571 68,991 5 0 69 27 85 2 4103 Web Mt. Web Mt. TN 26,389 8,854 34 1,190 5 0 0 95 21 23 Fort Mt.Rock Creek/TailsCreek/Cold Spring GA 27,04826,713 99 317 10,026 0 1 37 62 18 493 Hiwassee Lake Hiwassee Lake NC 14,891 9,865 66 8,519 0 0 57 43 11 52 18315,2510Average 42,131 2 54 1,440 79 21,884 2 0 41 57 34 9 613,85164 Mid-Chattooga Mid-Chattooga GA,SC 38,681 3 36 3,210 26,866 8 0 69 22 58 6 1235,82224 Lower Chattooga Lower Chattooga GA,SC 78,235 5 46 1,337 2,532 42,977 2 3 55 40 41 410,21204 Sumpter Sumter SC 34,911 0 29 501 22,420 1 0 64 34 32 3419,9619Average 50,609 3 37 1,682 844 30,754 4 1 63 32 44 7 4130,82 39,13315 Fairy Stone Fairy Stone VA8 3 30 4,732 2,781 0 4 2 94 51 5 1911,015 Bull Mountain Bull Mt. VA 21,804 7 51 389 2,249 0 2 10 88 4 43 23Pinnacles of42,55 24,210 105 Pinnacles of Dan DanVA0 10 57 0 0 0 0 8 5 6%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)41Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibiansHectares1.670971.8 1 1 2 1 45 1 8 17 5 23 6 166 97 0 79,2653.3153.2 3 3 2 9 12 37 5 3 15 5 69 70 0 15,7193.82 5 3 2 1 20 3 5 15 14 44.0 15,9092.919. 0.3 25 8.3 2.6 14. 3.6 83. 60.3 389.4 3 2 1 2 3 3 3 5 3 7 3 7 3 3 32.8123.6 1 2 2 1 8 4 1 8 101 29 0 41,3703.15 1 3 1 1 1 1 1 3 3.0 5,3836.45 2 4 2 1 1 1.0 4,5912.72 3 2 1 2 2 1 2 1 5 6.0 6,1823.80.7 2.21.225.0 2 3 25 5 1 5 0.5 2 5 9.25 33.3 14,3823.113321.6 2 3 1 2 18 2 7 6 1 51 10 96 75 0 40,5484.65 3 3 1 1 3 1 23 1 1 1 2 10 29.0 10,6792.50 1 2 2 1 1 1 1.0 10,9464.18 3 5 5 1 2 3 7 1 8 11.0 6,0263.61.7 0.2 13. 2.7 24.2 2 3 2 0 2 1.5 6.5 0.5 40 5 5 3 5 8 23.5 90.55.78 3 3 1 1 1 12 1 6 7 46 30 73.0 15,6543.3263.8 3 3 3 1 8 16 14 7 17 199 73 0 31,6616.7320.4 4 3 4 1 1 8 1 23 4 37 244 72 0 14,1285.35.316. 0.3 5.6 20. 58.3 218.0 3 3 3 1 03 63 3 7 3 163 3 72.80 8 3 3 1 52,9442.15 2 1 3 1 8,8242.617,217 4 2 1 19BirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences

ELU_GroupBlock Name Revised Name States Acres65,065 Average119,336 Doe Mt. Doe Mt. TN819,336 Average855,847 The Snake The Snake VA,TN6Bald Mts./Rocky97,477 Rocky ForkForkNC,TN158,987 Holstien Mt. Holsten Mt. VA,TN470,767 Average736,268 Max Patch Max Patch NC,TN836,268 Average8Great Smokies9WestLargest Roadless Block% of blockAcres GAP124,786 461,707 1,677 0 2 4 94 21154 1618,61066 97 2 0 0 0 0 19 36 018,61066 97 2 0 0 0 0 19 36 024,56,54172 44 5 41,225 2 0 74 14 43 75 786,37,961254 89 6 54,546 8 0 56 36 75 9 337,621 64 53,387 0 0 91 9 26 67 149,54,8316 66 7 49,719 7 0 73 20 48 90 412,41159 34 19,821 0 0 55 45 49 6 212,41159 34 19,821 0 0 55 45 49 6 2Great Smokies14West NCTN 340,182 313,035 92 297,838 3,104 88 0 1 12 8Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)42Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibiansHectares2.54 5 2 2 1 02.82 8 2 1 1 1 2 2 3 6 8.0 7,8262.82 8 2 1 1 1 2 2 3 6 8.02.11222,602 3 1 1 100 4 39 15 6 6 0 47 290.0 02.039,449 2 2 2 7 10 5 102 12 1 33 1 97 81 268.0 51.51123,877 1 1 1 45 3 10 12 1 2 5 8 44 196.0 01.911 57.3 251.3 2 1 1 2.33 51.7 4 50.3 13 0.67 13.7 4 2 3 34.514,673 5 3 1 1 2 12 2 6 3 23 28 49.0 74.5 53 % 3 1 1 2 12 2 6 3 23 28 49.0402072. 137,660 16 1.61 1 2 1 15 826 14 886 93 31 25 49 133 158 0 79 Cove Mt. WMA Cove Mt. WMA TN 13,600 13,494 99 7,838 58 0 0 42 29 18 3.45 2 8 4 5 4 1 3 2 19 27.0 5,5049 Brasstown Bald? Brasstown Bald GA 23,505 23,338 99 13,076 31 6,333 56 0 27 17 10 24 4 1.44 1 1 2 8 41 36 51.0 9,5129 Rich Mt. Rich Mt. GA 29,728 24,290 82 10,312 195 11,037 35 1 37 28 17 22 1 1.34 1 1 2 16 15 19.0 12,031JoyceKilmer/UnicoiJoyceKilmer/Unicoi16 832287. 153,319Mts./Slick Rock Mts./Slick Rock NCTN 378,846 84,087 22 37,485 215,162 10 0 57 33 6 2 16 2.64 2 3 1 1 12 424 32 1388 112 15 60 87 157 156 0 49 Cheoah Cheoah NC 39,676 38,558 97 225 23,479 0 1 59 40 85 77 7 4.08 2 3 1 4 19 4 1 4 1 4 2 17 29 56.0 16,0569 Fires Creek Fires Creek NC 60,413 45,199 75 41,342 0 0 68 32 48138 1 3.07 2 2 1 16 4 1 2 4 3 2 13 32.0 24,4489 Harmon Den Harmon Den NCTN 37,555 27,552 73 24,254 0 0 65 35 32108 2 3.74 3 3 1 8 1 3 6 1 6 19 21.0 15,198119 Yellow Creek Yellow Creek NC 40,684 14,687 36 22,209 0 0 55 45 72 1 11 4.47 3 4 1 1 26 2 5 2 12 3 5 20 30.0 16,4641951.69 Average 107,132 64,915 75 39,219 50 40,055 27 0 41 32 67 2 6 2.87 1 2 0 0 1 7.56 144 6.11 254 25.9 5.89 11.9 16.3 42.1 7 510.610 Nantahala Nantahala NC 16,621 15,597 94 10,061 0 0 61 39 43 32 4.56 1 3 1 2 1 1 1 1 2 5 12 15 25.0 6,72610 Average 16,621 15,597 94 10,061 0 0 61 39 43 32 4.56 1 3 1 2 1 1 1 1 2 5 12 15 25.011Great SmokiesEast Great Smokies East NCTN 254,768 197,392 77 194,895 960 10,250 76 0 4 19149526 2.65 2 2 1 30 255 32 5 109 10 20 76 155 147 692.0103,1011711 Plot Balsams Plott Balsams NC 48,520 31,847 66 1,714 1,940 5,716 4 4 12 81 65 1 0 4.85 2 4 1 17 3 1 5 4 13 19 30 62.0 19,6361311 Rabun Bald Rabun Bald GANC 49,824 23,149 46 1,092 41,948 2 0 84 14 56 5 10 3.83 1 3 1 4 14 1 22 1 1 13 45 55 101.0 20,163BirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences

ELU_GroupBlock Name Revised Name States AcresLargest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)11 Qualla Qualla NC 15,463 6,762 44 180 938 1 6 0 93 47 62 7.05 2 6 1 1 2 1 1 4 4.0 6,2581011 Black Mts. Black Mts. NC 71,338 36,193 51 2,635 47,725 0 4 67 29 6 94 5 2.80 1 2 1 47 10 32 47 2 34 26 97 90 295.0 28,8691711 Little Tennessee Little Tennessee NC 74,460 25,455 34 48,327 0 0 65 35 95 0 11 3.55 2 3 1 15 6 4 46 2 4 234 20 27 60 358.0 30,1331964.311 Average 85,729 53,466 53 32,980 1,079 25,661 14 2 39 45 86 3 4 4.12 2 3 0 18.8 48 11.5 9 30.8 3 49 24.7 57.2 3 252.012ForgeMt./RogersRidge?Forge Mt./RogersRidge43Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed PineNCVATN 37,112 25,745 69 170 16,424 0 0 44 55 26 53 2.15 2 1 1 6 2 17 1 3 24 20 53.0 15,01912 Nolichucky Nolichucky NCTN 33,877 30,515 90 26,017 0 0 77 23 49 39 3 2.60 1 3 3 51 1 43 5 1 45 5 69 61 220.0 13,71012 Average 35,495 28,130 80 85 21,220 0 0 61 39 38 46 2 2.38 2 2 2 28.5 1.5 30 3 0.5 24 2.5 46.5 40.5 136.51313a Linville Gorge Linville Gorge NC 72,650 27,624 38 11,701 645 36,777 16 1 51 32 34 3 2.30 2 3 4 2 1 24 8 1 5 1 6 17 64 52 126.0 29,401AdamsAdams Mt./13aMt./John's River John's River NC 19,189 16,402 85 10,506 0 0 55 45 51 28 2 4.14 4 2 3 1 1 1 8 7 17.0 7,76613a Globe Globe NC 35,589 14,609 41 17,716 0 0 50 50 51 71 3 3.42 3 2 3 1 1 3 1 2 2 7 9.0 14,40219.613a Average 42,476 19,545 55 215 25,567 0 0 57 43 46 77 1 3.29 3 2 3 1 0 8 3 0.67 3 0.33 5 6.33 24.3 7 50.713bSouth Mt.ConnectorSouth Mt.Connector NC 11,266 9,316 83 0 0 0100 3 17 1.78 2 1 2 1 1 3 4 4.0 4,55913bAverage 11,266 9,316 83 0 0 0100 3 17 1.78 2 1 2 1 1 3 4 4.0 4,55914a Poor Mountain Poor Mt. VA 45,353 39,287 87 2,630 33 4,061 6 0 9 85 0 95 2.09 2 3 1 1 18,35414a Average 45,353 39,287 87 2,630 33 4,061 6 0 9 85 0 95 2.09 2 3 1 11014b Macks Mountain Macks Mt. VA 42,006 17,250 41 10 0 0 0 0 68 22 1.62 2 1 2 1 16,99914bAverage 42,006 17,250 41 10 0 0 0100 68 22 1.62 2 1 2 114cHorsetroughMt.? Horsetrough Mt. GA31,872 25,070 7917,08712,38054 0 39 854 45333.12 2 3 1 1 33 18 38.012,89814c Blood Mt.? Blood Mt. GA20,206 9,585 47 7,733 40811,22138 2 56 436 33183.38 2 1 1 8 1 6 2 43 36 57.0 8,177Raven Cliffs Raven Cliffs27,0715,12 35 1 3.110,9514cWilderness? WildernessGA7 20,543 76 9,3384 4 0 56 10 6 28 8 2 1 2 1 1 1 3 1 19 13 25.0 846,5910,5134,97 25 10 2 2.718,8514c Unicoi Unicoi GA4 17,749 31 0 2,744 1 3 6 75 -4 5 2 4 0 1 3 1 1 8 6 16 3 58 29 91.0 614c Average31,437 18,237 5811,167 78818,42437 2 56 450 52233.08 1 2 1 1 52.25 6.50.750.7538.3 24 52.815aHoneycutMt./HumpbackMt.HoneycutMt./HumpbackMt.NC15,930 10,535 66 788 1,875 0 5 12 8351 26 1% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibians4.83 4 4 1 2 1 17 4 20 24.0 6,44715a Little Little NC 30,20 23,048 76 506 17,34 0 2 57 41 7 60 4.5 1 3 1 1 1 1 1 1 2 5 9 12.0 12,22BirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element OccurrencesHectares

ELU_Group15aAverageLargest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)Roads (4wd and trails)HectarBlock Name Revised Name States AcresesSwitzerland Switzerland 8 9 6 1 515b Elk Ridge Elk Ridge NCDoughton/Stone Doughton/Stone15bMt.Mt.NC15b Rendezvous Mt. Rendezvous Mt. NC15b Vannoy Vannoy NCYadkinYadkin15bHeadwaters Headwaters NC15bAverage15c Bent Mountain Bent Mt. VA15c Average16 Blue Wall Blue Wall NCSC16 Mt. Bridge West Mt. Bridge West NCSC16 Chimney Rock Chimney Rock NCHickory Nut Hickory Nut16GorgeGorgeNC16 Green River Green River NC16 Jocassee Jocassee NCSC16 Mt. Bridge Mt. Bridge NCSC16 Average17a Spring Creek Mt. Spring Creek Mt. NC17aAverageUpperUpperChattooga/Thre Chattooga/Thre GANCS17be Statee StateC17b White Water White Water NCSC23,069 16,792 71 647 9,612 0 3 35 6251,882 12,064 23 285 366 50 1 1 0 9938,0519,9253 20,529 544 6,626 0 2 17 3041,992 12,961 31 1,280 3,224 0 3 8 8945,355 11,770 26 558 91 0 1 0 9949,846 19,259 39 96 7,028 0 0 14 8645,4216 15,316 35 57 4,445 3,404 0 2 8 8031,92106 12,704 40 26 0 0 0 031,92106 12,704 40 26 0 0 0 056,3421,0830 17,582 31 7 131 1,357 7 0 2 6014,061 26 12,409 88 1,533 4,096 1 1 9 0 6013,7323 9,937 72 402 3,380 352 3 5 3 7065,899 26,028 39 1,544 745 3,048 2 1 5 9230,2113,029 15,719 52 1905 1 0 43 5682,2132,9446 45,097 551 8,778 0 0 11 4928,669 11,471 40 1,578 2,931 0 6 10 8441,5912 19,749 54 3,537 6,124 4,213 8 4 11 6740,590 20,039 49 7,107 0 0 18 8240,590 20,039 49 7,107 0 0 18 8258,880 11,052 19 4,186 22039,868 9,837 25 1,457 85738,455 7 0 65 2721,948 4 2 55 3964 43 05 12 11 2 264 81 055 74 462 94 553 89 457 92 512 6112 613 201 7 025 43 013 32 34 133 710 78 07 189 433 69 03 103 7 149 95 049 95 07161186 3147Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibians4.67 3 4 1 1.5 0.5 0.5 0.5 1 9.5 4.5 14.5 18.03.320,994 6 3 1 2 7 3 1 1 3 6 13 21.0 63.815,390 5 3 2 1 11 1 2 7 5 3 13 29.0 93.016,996 6 3 1 3 5 1 6 6.0 43.418,354 7 3 2 2 1 1 1.0 52.820,174 4 2 1 1 1 2 3 3.0 23.30 6 3 1 2 4.6 0.8 0.6 2 0.2 1.6 2.2 7.2 12.02.312,920 3 302.30 3 34.222,802 4 5 1 1 18 3 1 18 1 1 7 183 78 235.0 04.84 2 3 1 1 20 10 19 1 9 70 67 134.0 5,6923.28 4 3 1 22 17 4 2 1 5 6 66 44 123.0 5,5582.726,662 3 2 1 1 24 10 1 1 2 7 6 21 30 72.0 82.912,224 1 3 2 1 24 6 2 2 5 66 32 105.0 93.21277. 33,270 2 2 4 1 40 57 2 18 62 13 6 221 854 198 0 13.511,605 5 4 2 1 1 1 13 2 1 1 1 2 12 21 27.0 23.514.2.8 14.36.67.13 3 3 2 0 0 1 23 1 1.86 6 7 3 3 6 182 4 281.93.516,425 5 3 2 11 16 2 12 24 41.0 63.55 5 3 2 11 16 2 12 24 41.04.36 2 3 1 48 29 3 7 25 4 4 58 217 145 396.05.21 2 2 4 1 29 13 5 18 1 27 180 99 276.0BirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences23,82816,13444

ELU_GroupBlock Name Revised Name States Acres52,2113,80 11,08 24 1317b Headwaters Headwaters NCSC 2 36,876 71 1,677 9 8 3 6 21 49 7 2 215,6011,0917b Tiger Tiger GA7 15,431 994 0 0 71 29 9 4317b41,6420,644 12Average2 18,299 54 1,830 3,722 6 3 7 53 36 7 7 149,467,09 35,181317c Duncan Ridge? Duncan Ridge GA 9 15,226 31 3 7 6 0 14 71 15 60 8 713,3510,6717c Rocky Mt.? Rocky Mt. GA 3 12,927 97 194 6 0 1 80 19 35 2831,413,64 22,9317c Average1 14,077 64 1 6 1 0 8 76 17 47 83 390,702,42 8,90 57,2012 2018 Grandfather Mt. Grandfather Mt. NC 1 18,438 20 9 1 8 3 10 63 24 8 6 181818 AverageAshevilleWatershed/Curtis CreekAshevilleWatershed/Curtis CreekNCLargest Roadless Block% of blockAcres GAP172,746 44,181 61 14381,721,284 31,309 41 61,4355,168Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)36,466 0 2 50 48 71 94 346,83157 1 6 57 36 99 0 1019a Dupont West Dupont West NC 9,826 7,802 79 171 2,728 2 0 28 70 6 3625,1410,7419a Dupont Dupont NC 4 9,103 368 0 0 43 57 9 64 119a17,48Average5 8,453 58 86 6,738 1 0 35 64 7 50 025,8018,7519b Cathy's Creek Cathy's Creek NC 8 10,287 407 0 0 73 27 59 43N. Fork French N. Fork French14,5519bBroadBroadNC 9 4,873 33 7,269 0 0 50 50 42 5719b20,1813,01Average3 7,580 373 0 0 61 39 50 5069,164,5445,961219c Mount Rogers Iron Mt. East VA 6 17,659 26 9 28 0 7 0 66 27 20 6 869,164,5445,961219c Average6 17,659 26 9 28 0 7 0 66 27 20 6 851,991,8437,6320a Chunky Gal Chunky Gal NC 3 25,712 49 27 4 0 72 24 70 9663,6420,752520a Cowee Cowee NC 2 13,208 214 0 0 33 67 88 868,6114,792420a Tri-County Tri-County NC 9 24,190 354 0 0 22 78 56 420a61,4124,3919Average8 21,037 35 6145 1 0 42 57 72 969,806,9626,481820b Pond Mt. Pond Mt. NCTN 0 12,865 18 43 10 0 38 52 49 6 1645Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibiansHectares3.421,122 2 3 1 1 17 20 5 2 46 2 1 39 300 139 436.0 93.35 2 3 3 1 1 1 1.0 6,3164.015.22. 1.8 2 3 2 1 23.5 5 2 3.5 3 5 1.5 31 175 96 277.34.020,010 1 2 1 1 5 1 10 5 1 3 53 29 79.0 94.74 2 2 1 1 2 2 11 8 15.0 5,4044.31.7 2 2 1 1 0.5 3.5 1.5 5 2.5 0.5 5 32 18.5 47.03.6158. 36,708 3 3 1 1 22 2 6 25 5 9 19 70 63 0 52.26 1 2 54 9 2 2 7 9 21 38104.029,4392.913.45.131.7 2 3 1 1 38 1 7.55 3.5 8 14 5 50.5 04.28 3 3 1 1 9 8 1 39 19 57.0 3,9772.9132. 10,171 3 3 3 1 50 28 1 1 4 48 31 0 53.543.9 3 3 2 1 29.5 18 0.5 1 2 5 25 94.53.910,445 1 2 1 2 3 3 3 1 6 12 18.0 46.83 5 6 1 3 3 1 3 1 6 12 17.0 5,8925.39 3 4 1 2.5 3 2 3 1 6 12 17.52.127,991 3 2 4 112.11 3 2 4 13.1127. 21,049 2 2 1 9 26 2 3 8 10 14 55 56 0 15.4134. 25,755 3 4 1 37 2 1 2 1 6 5 18 62 55 0 54.327,767 3 3 2 6 2 2 1 1 1 15 19 28.0 94.30.6 1.643.34 3 3 1 17.3 10 1.67 7 7 5 5 11 44 3 96.33.3350. 28,247 5 4 2 7 5 43 8 1 15 6 265 57 0 7BirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences

ELU_GroupLargest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%UnsecuredRoads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)HectarBlock Name Revised Name States AcresesUnika Mt. Unaka Mt.58,024,4617,752.9229. 23,4820bWilderness Wilderness NCTN 9 25,745 55 58 8 0 31 62 56 83 1 8 4 2 2 4 73 6 8 59 14 2 64 67 0 439,7511,304.316,0820b Hurricane Mts. Hurricane Mts. NC 7 16,183 419 0 0 28 72 72 990 7 3 1 1 11 1 2 1 1 18 20 34.0 949,6511 2.920,0920b Big Laurel Big Laurel NCTN 3 20,321 41 5,100 0 0 10 90 33 2 3 4 5 3 2 2 9 1 5 3 10 20.0 420b54,312,8515,1612 3.322.0.2 8.7 87.158.Average0 18,779 39 72 4 0 27 69 53 0 5 9 5 3 0 2 4.25 5 3.25 13 17 5 5 2 5 38.5 3Jefferson/Iron75,889,38 4,56 47,9714 2.230,7021Mt. Mt. Rogers VA 4 34,681 46 2 9 3 12 6 63 18 31 17 2 1 1 4932,1820,042.113,0221 Sugar Grove Glade Mt. VA 4 8,369 261 0 0 62 38 7 62 6 5 2 2 3454,034,69 2,28 34,0010 2.221 Average4 21,525 36 1 5 7 6 3 63 28 19 2 3 1 2 2 1 3.555,221,49 18,0413 3.7595. 22,3422a Roan Mt West Roan Mt West NCTN 5 19,463 35 686 2 2 1 3 33 63 74 5 4 9 5 3 1 1 20 149 17 25 48 53 51 232 143 0 932,523.413,1622a Big Bald Big Bald NCTN 4 19,395 60 6,498 0 0 20 80 45 68 1 7 5 3 1 2 4 14 1 6 1 29 37 57.0 222a43,8712,2710 3.675.19. 24.29.326.Average5 19,429 48 343 746 0 1 1 26 72 59 2 3 3 5 3 1 1 10 5 10.5 5 55 26 131 90 0Standing Standing95,5011,6269,1512 1.7246. 38,6522bIndian Indian GANC 8 57,585 60 6 315 5 12 0 72 15 43 7 15 8 1 1 1 34 29 5 9 19 7 2 11 130 102 0 157,3045,731.9125. 23,1822b Pink Beds Pink Beds NC 0 31,636 55 646 5 0 1 80 19 52 594 1 2 42 16 1 1 3 2 1 5 54 39 0 932,9715,6111 4.3148. 13,3422b Cullasaja Cullasaja NC 4 25,535 774 0 0 47 53 30 23 2 3 1 38 1 3 5 9 53 39 69 0 4Panthertown Panthertown49,6513,0716 4.6214. 20,0922bValleyValleyNC 8 13,996 284 0 0 26 74 63 6 0 2 4 3 2 74 5 1 5 5 55 69 63 0 658,8635,8911 3.112.6.2 4.7 4.268.2 183.22b Average0 32,188 55 2,906 240 4 3 0 56 40 47 6 4 7 2 2 1 47 5 2 2.5 5 5 5 31 73 5 314,205.223 Mt. Pisgah Mt. Pisgah NC 2 8,194 58 690 3,163 0 5 22 73 31 43 0 2 3 4 1 7 1 4 3 4 8 25 27.0 5,74793,621,64 44,8811 28 4.3202. 37,8823 Roy Taylor Roy Taylor NC 5 40,730 443 2 0 2 48 50 9 3 0 0 3 3 1 59 3 5 15 17 1 2 17 83 60 0 920,4319,213.623 Looking Glass Looking Glass NC 9 17,234 84 350 7 0 2 94 4 53 212 1 3 1 21 4 1 12 41 21 79.0 8,27142,7522,4211 4.30.335.3 102.23 Average5 22,053 62 894 1 0 3 55 42 68 6 0 8 2 3 1 29 1 3.33 5 7 3 2 11 44 3 7Richland Richland30,701,03 24,524.112,4224Balsam BalsamNC 4 18,049 595 3 0 3 80 17 69 583 1 2 1 26 1 7 19 1 8 25 43 87.0 535,8126,964.314,4924 Shining Rock Shining Rock NC 8 31,825 89 581 8 0 2 75 23 78 808 1 2 2 9 1 10 42 1 1 9 14 44 87.0 533,2625,744.230.19.24 Average1 24,937 74 808 6 0 2 78 20 73 696 1 2 2 17.5 1 8.55 0.5 1 8.5 5 43.5 87.0Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)CommunitiesAmphibiansBirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences46

ELU_GroupBlock Name Revised Name States AcresAmphibolite Amphibolite51,0225Mts.Mts.NC 3 16,528 32 3,00165,5825 Roan Mt. East Roan Mt. East NCTN 1 38,340 58 1,71458,3025 Average2 27,434 45 2,358Largest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 3%GAP 1%GAP 2%GAP 3%Unsecured2,696 3,659 6 5 7 82 61,152 7,506 3 2 11 84 501,924 5,582 4 4 9 83 28Roads 1-5 (Hwy & Routes)Roads 6-8 (Local roads)108153 0131 0Roads (4wd and trails)Road density per 1000 acres% Pasture/Hay% % Developed Open SpaceEvergreen Plantations or Managed Pine% Row Crop% Low Intensity Developed% Clearcut - Grassland/Herbaceous% Successional Shrub/Scrub (Utility Swath)% Successional Shrub/Scrub (Other)% Successional Shrub/Scrub (Clear Cut)Communities2.24 5 2 2 93 5 16 3 2 7 2 32 262 1283.10 6 3 2 20 125 12 26 51 4 7 13 179 962.614. 26.22.7 6 3 2 56.5 65 14 5 5 5.5 4.5 5 221 112AmphibiansBirdsFishMammalsReptilesInvertebratesNon-Vascular PlantsVascular PlantsTotal TaxaTotal Element Occurrences422.0437.0429.5Hectares20,64826,540GrandAverage53,822 26,774 55 7,4111,57121,675 7 3 40 50 48117 53.45 3 3 1 0 0 0 1 0 0 12 26 2.9 37 11 1.9 8.2 12 62 40.7171.847

Appendix F: Key to Columns in Appendix EELU_GroupBlock NameRevised NameStatesAcresField NamesLargest Roadless Block% of blockAcres GAP1Acres GAP 2Acres GAP 3DescriptionA group of forest blocks with similar geophysicalattributes. (The number is derived from a cluster analysis)Original name assigned at the block meetingsRevised namesStates included in this block%GAP 1 % of block in GAP 1%GAP 2 % of block in GAP 2%GAP 3 % of block in GAP 3%UnsecuredRoads1-5 (Hwy & Routes)Roads6-8 (Local roads)Roads(4wd and trails)Block size in acresThe size of the largest completely roadless block inside theoverall blockPercent of forest block covered by the largest roadlessblockAcres of GAP 1 secured land (permanent protection fornature)Acres of GAP 2 secured land (permanent protection withsome manipulations)Acres of GAP 3 secured lands (permanet securement formultiple uses)% of block that is unsecuredMiles of roads class 1-5 (interstates, US routes, Stateroutes, Parkways, service roads, ramps)Miles of roads class 6-8 (local roads lanses streetsavenues, driveways)Miles of 4 w drive roads and trailsRd density per 1000 acres Miles of class 1-8 roads divided by block size * 1000% of modified environments based on the SE GAP habitatPasture/Haymap% of modified environments based on the SE GAP habitatDeveloped Open SpacemapEvergreen Plantations or ManagedPine (can include dense% of modified environments based on the SE GAP habitatsuccessional regrowth)map% of modified environments based on the SE GAP habitatRow Cropmap% of modified environments based on the SE GAP habitatLow Intensity Developedmap% of modified environments based on the SE GAP habitatClearcut - Grassland/Herbaceous mapSuccessional Shrub/Scrub (Utility % of modified environments based on the SE GAP habitatSwath)map% of modified environments based on the SE GAP habitatSuccessional Shrub/Scrub (Other) map48

Field NamesSuccessional Shrub/Scrub (ClearCut)CommunitiesAmphibBirdsFishMammalsReptilesInvertsNon VascularVascular PlantsTotal TaxaTotal EOHectaresDescription% of modified environments based on the SE GAP habitatmapTracked Community occurrencesOccurrences of tracked amphibiansOccurrences of tracked birdsOccurrences of tracked fishOccurrences of tracked mammalsOccurrences of tracked reptilesOccurrences of tracked invertebrates (insects, mussels,spiders etc)Occurrencs of non vascular plants (mosses, lichens, ferns)Occurrences of vascular plantsTotal number of speciesTotal number of occurrencesBlock size in hectares49